Stratos Family OS

Every card is empty. Fill them with what AI could actually do inside these companies.

Forty nine real organisations, one card each. Pick one, work out what the people inside it could stop doing by hand, and write it on the card. Not one use case. As many as you can see. For at least one of them, say what five minutes on a screen would look like.

Book a 30-minute call
0use cases on the wall
0people who have added one
49cards still empty
How to add yours, from your terminal

You have push access. No pull request, no waiting for anyone.

  1. Clone it: git clone https://github.com/Stratos-Technologies-fzco/stratos-family-os.git
  2. Open docs/index.html and search for the company name. Every card sits between <!-- ===== START name ===== --> and <!-- ===== END name ===== -->.
  3. Paste the block below inside that card's <div class="cases">. Add as many as you like.
  4. git pull --rebase, then git add -A, git commit -m "yourname: use cases for X", git push.

The page rebuilds in about a minute and your use case is on the wall with your name on it. Edit inside a card only. Everything outside belongs to somebody else.

The block to copy. Keep the four lines. The last one is your GitHub name.

<div class="case">
  <h3>The weekly field report writes itself</h3>
  <p><b>Today</b> a field officer visits twelve schools a week and writes the report by hand every Friday. Three hours, and most of it is the same every week.</p>
  <p><b>With AI</b> an agent reads the visit notes and last week's report and drafts this week's. Friday becomes twenty minutes.</p>
  <p><b>Five minutes</b> open three real visit notes, run it live, read the report out, change one note and run it again so they see it is not a template.</p>
  <p class="by">@yourhandle</p>
</div>

What makes one good. Name the person it happens to. Say what they do today and how long it takes. Then show something moving on a screen. "AI chatbot for customer service" is not a use case. If you are guessing about how a company works, say so in the card.

Never write anything about prices, contracts or private conversations here. This page is public.

Advaith Foundation

Education nonprofit · Bangalore, India

What they do

A charitable trust set up in 2004 that pays for the entire schooling, food and housing of children from very poor families in and around Bangalore, free to the family. Its Ankur programme puts those children into small residential homes and sends them to The Samhita Academy, a CBSE school off Bannerghatta Road, where they sit in the same classrooms as children whose parents pay fees; the money comes from donors and company CSR budgets, not from the children.

Inside, day to day. Teachers plan and mark lessons and run after-school tuition, house parents and wardens live with the children and track each one's attendance, health, marks and behaviour largely on paper, and a small office team writes the donor and CSR reports, keeps the trust's compliance filings in order, and handles the visits and paperwork that come with taking in a child.

The scholarship form stops living as a photo

Today each Grade 5 scholarship application to Ankur arrives by post, email or WhatsApp, as a photo or scan of a form with a passport photo, Aadhar copy, birth certificate, caste and income certificate and last report card attached; an administrative staff member opens it, checks the set is complete, reads the declared family income against eligibility, and then — and this last step is a guess, since no online application route exists on the site — copies the details by hand into the list the intake runs from, one applicant at a time, all through the intake window.

With AI she builds and runs a tool that reads each submitted form — scan, WhatsApp photo or emailed PDF — pulls out the child's details and the declared income, checks the five required documents are all there, and drops a completed row into the intake list, flagging anything missing for her to chase; she keeps tuning it herself as the form's fields change each admissions cycle. That is time back for the eligibility judgement itself, which stays hers.

Five minutes open three synthetic Grade 5 applications built to match the real form — one a WhatsApp photo, one a scanned post submission, one an emailed PDF — run the tool live and watch a structured row and a missing-document flag appear for each; then, live, change the declared annual income on one synthetic form and re-run it, so the audience watches the eligibility flag flip because it read the new number, not a cached one.

@sahidmodi45

The disclosure pack builds on last year's, not from zero

Today once a year the Principal's office pulls together The Samhita Academy's Bangalore CBSE mandatory disclosure pack — eight uploaded documents plus five tables covering fee structure, academic calendar, SMC and PTA lists and three years of board results — and refreshes every figure in it; the version filed against the 2022-23 calendar and the current 2026-27 version are laid out completely differently, which — and this is a guess, since no system is named for it — suggests each year's pack gets rebuilt rather than carried forward and edited.

With AI the office builds a tool that reads last year's filed pack alongside this year's new figures and drafts the refreshed eight-document, five-table pack in one consistent structure every time, so assembling it becomes a check-and-correct pass they run themselves instead of a rebuild from a blank file. That is time back before every filing deadline.

Five minutes open the school's own two published disclosure PDFs — the 2022-23 pack and the 2026-27 pack — run the tool live to show it mapping the older pack's figures into the newer structure; then, live, change one figure in a synthetic 'this year's update' sheet and re-run it, so the audience watches that one table update on screen.

@sahidmodi45

<

The discovery interview turns into the process diagram

Today a consultant spends hours in discovery interviews with a client's finance and operations teams, taking running notes, and then spends another full day manually mapping the steps into workflow diagrams and integration specifications before engineers can write a line of code.

With AI an agent reads the raw interview transcript and drafts the step-by-step process flow and data-handover schema in standard diagram notation, leaving the consultant to verify edge cases and business rules rather than drawing boxes from scratch — that is time back.

Five minutes open a synthetic transcript of a 30-minute discovery call with a mid-market finance team, run the agent live to watch the workflow diagram and field-mapping table generate on screen, then edit one line in the transcript to add a manual approval step and run it again so the room sees the diagram update its logic live.

@saiqulmodi

>

AdySphere

Enterprise systems, automation · United Kingdom and UAE

What they do

A small consulting firm, founded by Adyasha Mallick, that builds the internal systems a company runs on: management dashboards and reporting, ERP and CRM setups, and automation that carries data between departments so nobody has to retype it. Its customers are mid-sized and larger companies in the UK and the Gulf that already own a lot of software and want it mapped, joined up and made to run without manual handoffs.

Inside, day to day. Consultants sit with a client's finance, operations or sales team to work out how the work actually flows today and draw it out, and then engineers build the dashboards, data pipelines and integrations that replace the spreadsheets, so the days go on discovery interviews, process diagrams, and writing and testing integration code against somebody else's systems.

Nobody has written a use case for this one yet.

Al Ain University

Private university · Al Ain and Abu Dhabi, UAE

What they do

A private university with two campuses in the UAE, teaching roughly 5,000 students from more than twenty nationalities across eight colleges: engineering, pharmacy, law, education and humanities, business, communication and media, dentistry, and nursing. It sells degrees from bachelor's through doctorate, and its customers are students and the families who pay the tuition.

Inside, day to day. Around 370 faculty and staff spend their days lecturing, setting and grading coursework and exams, supervising masters and PhD students through their research and thesis writing, publishing their own papers, and producing the accreditation and quality-assurance evidence that UAE regulators demand of a licensed university every year.

The same admissions question, answered fresh each time

Today a Student Recruitment Officer at the Al Ain campus answers the same admissions questions — where an application stands, what it still needs — by email and phone, one enquiry at a time, through the cycle that brought in 1,663 new students last fall; how many enquiries land in a day could not be confirmed, only that the number of questions runs well above the number of students who actually enrol.

With AI she builds a tool that reads the university's own published list of its eight colleges and 49 programmes and drafts the reply to each enquiry from that, and she is the one who edits it herself when a new intake changes the details — freeing her time for the applicants who need an actual conversation, not the same status question forty times over: that is time back.

Five minutes open a synthetic inbox of three made-up student questions modelled on the recruitment officer's own duty line, load AAU's real public college-and-programme list, and run the first question — which college handles a Nursing application — reading out the drafted answer naming the College of Nursing. Then, live, edit the question to ask about Pharmacy instead and run it again, so the room sees the draft change to the College of Pharmacy rather than repeat the first answer.

@sahidmodi45

Altinad Partners

Capital advisory, finance · New York and Dubai

What they do

A small advisory firm that stands between people who need money and people who have it. It gets fund managers and company founders ready to raise capital, brings deals to family offices and institutional investors, and sets up the legal vehicles (SPVs, syndicates, co-investments) that let several investors go into one deal together, charging retainers and fees rather than selling a product.

Inside, day to day. A handful of partners and associates spend their days on calls with fund managers and family offices, reading decks and financial models to judge which deals are worth passing on, writing investment memos and assembling data rooms, and keeping track of which investor has seen which opportunity and what they said back.

The pass-or-hold call stops living only in an inbox

Today an Associate at the eight-person firm reads every inbound deck and financial model against whichever of the firm's nine published sector mandates it is pitched under — Artificial Intelligence, Digital Infrastructure, Venture, Growth Equity, Private Credit, Secondaries, Energy, Healthcare, FinTech or Cross-Border — and decides, deal by deal, what moves to diligence. Altinad names no screening system anywhere on its site, so we are guessing this filtering, and the reasoning behind each call, still lives in someone's inbox rather than anywhere searchable.

With AI she builds a screening assistant that reads each inbound deck and model against the specific mandate for whichever sector it is pitched under, and drafts the pass-or-hold note with the reasoning attached instead of a bare verdict, tuning the criteria herself as mandates change. More decks cleared to the same standard in the same week — that is productivity.

Five minutes open two made-up decks, labelled invented, one pitched under Digital Infrastructure and one under Healthcare, run the assistant against the Digital Infrastructure mandate and read out the pass-or-hold note it drafts, then change that deck's sector tag on screen to FinTech and run it again so they watch the verdict and the reasoning change with it.

@sahidmodi45

One data room becomes the template for the next

Today an Associate rebuilds the investor narrative and data room, and the mirrored structured diligence pack for the investor side, from scratch for every mandate that comes in — the same structure, tied to the firm's own six published process phases, redone with only the sector and vehicle changing. No drafting or data-room platform is named anywhere on the firm's site.

With AI she builds a drafting agent seeded with the firm's own six-phase structure that pulls in each new deal's sector, vehicle type and terms to draft both the sponsor narrative and the mirrored investor diligence pack, and she edits the deal-specific paragraphs herself instead of starting from a blank document each time. The same two-sided document set goes out for more mandates in the same week — that is productivity.

Five minutes open a made-up deal brief, labelled invented, for a Single-Deal SPV pitched under Artificial Intelligence, run the agent and read out the narrative and diligence-pack index it drafts, then change the vehicle type on screen from Single-Deal SPV to Syndicate and run it again so they watch the structuring section of both documents update to match.

@sahidmodi45

Who saw which deal stops living in someone's memory

Today — and this is a guess, built from the role and the firm's published commitments rather than anything describing the work itself, because the two intern postings that would likely confirm it could not be read — an Investor Relations Associate at the eight-person firm is likely keeping track, deal by deal and investor by investor, of who has seen which curated opportunity, what they said back, and what is owed to them next, with two more interns now being hired into exactly this function.

With AI she builds a tracker that logs, per investor and per deal, what was shown and what they said, and drafts the next update that is owed instead of piecing it together from inboxes; the two incoming interns learn the investor relationships on top of a system that already remembers the history, rather than spending their first months rebuilding it by hand — that is skill, on top of the same headcount.

Five minutes open a made-up list of three investors, labelled invented, each shown a different deal, run the tracker and read out what is owed to each, then type in a new reply on screen from one of them — "passes, wants to see the next Healthcare deal" — and run it again so they watch that investor's "owed next" line update.

@sahidmodi45

The investor update that never has a closing date drafts itself first

Today — and this part is a guess, because no reporting or fund-administration platform is named on the firm's site — an Investor Relations Associate assembles the ongoing report for each investor in each of the firm's four vehicle types by hand, for as long as that vehicle stays in the open-ended Administration phase.

With AI she builds a reporting drafter that pulls each vehicle's current numbers and drafts the update for that investor, and she writes the commentary herself where the story actually changed rather than rebuilding the report layout every cycle. The same associate can carry more open vehicles at once without the reporting backlog growing — that is productivity.

Five minutes open made-up numbers, labelled invented, for a Syndicate and a Co-Investment vehicle, run the drafter and read out each investor's update, then change the Syndicate's committed-capital figure on screen and run it again so they watch that investor's report sentence update to match the new number.

@sahidmodi45

Bharat Sikka Studio

Photography, film studio · New Delhi, India

What they do

A studio built around one photographer, Bharat Sikka, who makes fine-art photography shown in galleries and published as books, and who also directs and shoots advertising campaigns. He has two sets of customers: galleries, publishers and collectors on the art side, and brand and agency marketing teams on the commercial side, who have included Spotify, IndiGo, BMW, Royal Enfield and Tinder.

Inside, day to day. A tiny core team hires freelance crews job by job, so the days split between preparing shoots (locations, permits, casting, crew, kit), shooting, and then long stretches of selecting and retouching images, sending selects back and forth with clients and galleries, and keeping twenty-five years of archive, captions and image licensing straight; he also publishes a photography magazine, Neither, with its own submissions, editing and printing to run.

Nobody has written a use case for this one yet.

Cognitive Prism

Behavioural science lab · Bengaluru, India

What they do

Cognitive Prism Consulting Pvt Ltd says it measures how a person thinks, learns and makes decisions, turns those measurements into a written profile, and then gives advice and ongoing coaching off the back of it. It works in three areas, schools and universities, sport, and social programmes, so the buyers are institutions and teams wanting profiles of their students or athletes rather than individuals buying a test. Note: this is the Bengaluru company at cognitiveprism.in; several unrelated outfits use the name Prism, and the match has not been fully confirmed.

Inside, day to day. A team of roughly 11 to 50 people, still hiring research, sales and marketing roles, runs assessments on students and athletes, with analysts turning raw scores into a readable written profile for each person, counsellors and psychologists taking those profiles into one-to-one sessions, and a business-development side building a pipeline of schools and sports bodies by hand.

We are not certain we have the right organisation here. Check before you write.

Nobody has written a use case for this one yet.

CoLab Global

Social-impact consulting · Bengaluru, India

What they do

A consulting firm, founded and led by Rucha Pande, that works for nonprofits, foundations and government-facing education programmes in India. It is hired to answer whether a programme is really working and how to make it bigger: running research studies and needs assessments, designing the programme and its measurement framework, co-running small pilots, and then building the toolkits and open-source platforms that let the programme be copied by somebody else.

Inside, day to day. Consultants spend their days in the field and on video calls interviewing teachers, head teachers, field staff and programme leads, then come back and turn hours of notes and survey data into theory-of-change diagrams, evaluation reports and decks for the funder, so writing, rewriting and formatting long documents is most of the actual job.

Nobody has written a use case for this one yet.

Curio Lifestyle

Organic food brand · Jammu, India

What they do

Curio Lifestyle buys honey, ghee, cold-pressed oils, nuts, spices and tea from small farmers in Himalayan villages, packs them under its own label, and sells them online and in bulk. Shoppers buy directly from its Shopify website and from Amazon, and shops, grocers and wellness brands buy the same products wholesale.

Inside, day to day. It is a very small founder-run operation: the same handful of people arrange buying from farmers and self-help groups, get the goods bottled and labelled, then spend their days running the Shopify store, answering customer messages, keeping the Amazon and IndiaMART listings current, packing and tracking courier shipments, and chasing the supermarket and wholesale buyers who owe them money.

The "where is my order" message answers itself

Today one of the founders stops packing several times a day to answer customer messages from the Shopify store and Amazon — where is my order, is this honey raw, how long does the ghee keep — looking up the courier tracking and the product page by hand each time. I am guessing at the mix of questions; the card only says answering messages takes up part of every day.

With AI an assistant reads each incoming message, looks up that customer's order and courier status and the product's own listing, and drafts the reply in the brand's voice; the founder reads it and presses send, and edits the product notes it answers from whenever a new batch or product comes in. Anything about refunds or complaints is flagged for a person, never answered on its own.

Five minutes open five synthetic customer messages and a synthetic order sheet on screen, labelled as invented, run the assistant and read the five drafted replies out, then change one order's courier status from "in transit" to "delivered" and run it again so the room sees that reply change to match.

@shaoibul0926

One product change updates Shopify, Amazon and IndiaMART together

Today when a pack size, description or stock level changes, somebody edits the same product three times, once on Shopify, once on Amazon and once on IndiaMART, and the listings drift apart whenever one gets missed.

With AI the Shopify listing becomes the single source, and a tool writes the matching Amazon and IndiaMART versions from it, reworded to each site's format, and shows a list of what is out of step before anything is changed. The founder approves the list; nothing goes live without a click.

Five minutes show one synthetic product on screen in three columns, one per site, with one column deliberately out of date, run the checker so it highlights the mismatch and drafts the fix, then change the pack size in the Shopify column and run it again so both other columns update.

@shaoibul0926

Nobody has written a use case for this one yet.

Delta

Livelihoods incubator, nonprofit · Delhi, India

What they do

Delta (written "the^delta") helps small Indian nonprofits and social enterprises that are trying to create jobs and income for poor people — farm workers, women in informal work, young people looking for their first job. It runs open competitions to find promising ideas, then takes the best of them through an incubator and an accelerator, giving them money, advice and introductions; philanthropic funders and companies pay for the programmes.

Inside, day to day. The staff are programme people rather than field workers: they read piles of applications from small organisations, run selection panels, sit in coaching and review calls with founders, collect numbers back from every organisation in the cohort, and write the progress reports the funders expect.

Nobody has written a use case for this one yet.

Drishtee / Quiver Catalyst

Rural livelihoods, India · Noida, India (works across Bihar, Uttar Pradesh, Assam, Chhattisgarh, Karnataka)

What they do

Drishtee sets up small businesses inside villages — training centres, banking kiosks, women's production groups making food, textiles and handicrafts — and connects what those villages make to buyers in the cities. Quiver Catalyst is the money side of the same family: it puts investment and loans into rural enterprises that are ready to grow, and runs training with them.

Inside, day to day. Most of the real work happens in villages: field staff hold community meetings, recruit and hand-hold village entrepreneurs, run training camps, staff banking kiosks with micro-ATMs, collect goods made by women's groups and move them to urban buyers, and then fill in attendance, sales and beneficiary records on paper and phone forms that head office has to add up.

The CSP call log writes its own report

Today a call centre executive logs every inbound and outbound call to a Customer Service Point kiosk operator by hand into that CSP's profile record, then compiles those entries into a report for stakeholders when the reporting cycle comes round.

With AI he builds a call-note assistant that turns his notes from each call into the CSP profile update and flags anything a stakeholder needs to see, and when the reporting cycle comes round the report drafts itself from the profiles he has already checked. More calls turned into a usable record in the same day — that is productivity.

Five minutes open three made-up CSP call notes on screen, run the assistant, watch the profile fields and the draft report fill in, then read out a new call note live — a kiosk running low on cash — and run it again so they see that flag appear in the redrafted report.

@sahidmodi45

The cluster manager's daily report reads yesterday's first

Today a cluster manager makes his monthly round of CSP kiosk visits and drops in on the bank's local and regional head offices, then writes up a report at the end of each day and sends it up the chain.

With AI he types a short note after each visit into an agent he set up, which drafts the day's report against yesterday's so a CSP that is slipping, or a bank office that needs a follow-up, stands out; he edits and sends it himself. The end-of-day write-up goes from a chore to a five-minute check — that is time back.

Five minutes open two made-up days of visit notes — one CSP kiosk, one bank regional office — on screen, run the drafter, read out the report it produces, then add a new line to today's note live, "CSP #4 out of cash by 2pm", and run it again so they watch that line turn into a flagged follow-up in the redrafted report.

@sahidmodi45

The kiosk paperwork gets checked before the trip, not at the counter

Today a district executive appoints new kiosk operators in his area, collects each one's account-opening paperwork, and travels to the partner bank's office to check it against what the bank needs and hand it in.

With AI he photographs the paperwork before he leaves and an agent he set up checks it against the bank's own requirements for that appointment, flagging what's missing so nothing incomplete makes it to the bank counter; he still makes the trip and hands the documents over himself. Fewer trips spent fixing the same appointment twice — that is time back.

Five minutes lay out a made-up kiosk-operator document set on screen, run the checker, watch it list what's missing against the bank's requirements, then swap in a replacement document live for the one it flagged and run it again so they see the flag clear.

@sahidmodi45

GritWorks.ai

AI data privacy software · Bengaluru, India

What they do

GritWorks sells software that strips personal details out of a company's documents, images and audio recordings, and separately generates realistic fake documents and records that never belonged to a real person. Banks, insurers, hospitals and law firms buy it so their teams can build and test AI systems without passing real customer files around.

Inside, day to day. A small engineering team builds and tunes the models that spot sensitive information, makes the software run inside each customer's own systems rather than on GritWorks servers, and produces the audit trails a compliance officer will later inspect, while the founders spend their days in front of security, privacy and AI teams at large regulated companies. The same company earlier built conversational AI for collecting customer feedback, so both threads may show up in what you read about them.

Nobody has written a use case for this one yet.

Groupe Eram (GBO Asia)

Apparel footwear sourcing · Dongguan and Shanghai (China), Hong Kong, Dhaka (Bangladesh); parent group in France

What they do

Groupe Eram is a French family company that has been making and selling shoes and clothes since 1927 and owns about ten retail brands. GBO is its buying office in Asia: it finds the factories, develops the products with them and manages the suppliers who actually manufacture the footwear and clothing those brands sell in Europe.

Inside, day to day. The day is merchandising and quality control across four offices: staff chase factories for samples, prices and delivery dates, walk inspection lines checking stitching, materials and sizing, keep order and shipment trackers up to date, and email back and forth across time zones between factories in China and Bangladesh and brand teams in France.

Nobody has written a use case for this one yet.

Hello Chef

Meal-kit subscription · Dubai, UAE

What they do

Hello Chef sells a weekly subscription box of pre-measured ingredients and printed recipe cards to households across the UAE. You choose your recipes on their website each week, a chilled box arrives at your door, and you cook the meals yourself at home.

Inside, day to day. Chefs write around forty new recipes every week and the whole company works backwards from that: buyers order the exact quantity of each ingredient, a kitchen and warehouse team butchers meat in-house, weighs out every portion and packs the boxes, drivers run chilled delivery routes on fixed days, and a customer service team handles recipe swaps, skipped weeks, missing items and refunds.

The extras list gets checked before the box is packed

Today each week's box has to combine that week's core recipes — around forty of them, fully rotating every Saturday — with whatever extras a household has added, up to ten of them once it is past its first delivery, and, we are inferring this next part since no pick-list or warehouse system turned up anywhere in what Hello Chef has published, an Operations Packer works out each order's exact build by checking it against that week's list before it goes into the chiller.

With AI the packer builds a checklist tool that reads the week's confirmed orders and each household's extras and prints the exact build for every box, flagging anything it cannot match so the packer checks only those by hand instead of the whole list. More boxes go out right the first time in the same shift — that is productivity.

Five minutes open a synthetic Hello Chef order screen built from the real published forty-recipe rotation and allergen labels, with invented household names and extras, run the checklist tool and watch the build sheet appear with one allergen flag raised, then add a second extra to that same order live and run it again so they watch the build sheet and the flag both change with it.

@sahidmodi45

Hippocampos

AI memory software

What they do

Hippocampos (written "hippocampOS") sells a low-cost monthly subscription that connects to your email, notes, meeting records and files and turns them into a private, searchable memory. You give each memory a name — a work brain, a personal brain, a client brain — and then call it up inside Claude or ChatGPT so the AI already knows your context instead of you re-explaining it every time.

Inside, day to day. Judging only from the product, this is a small software team: they build and maintain the connectors that pull in Gmail, Notion, meeting notes and files, run the nightly syncing and the search that pulls the right slice of a person's history into a chat, and support paying subscribers. We could not establish who runs the company, where it is based, or how many people work there — the website names no founder, city or company details, so treat everything except the product itself as unknown.

Nobody has written a use case for this one yet.

IBPG

Business chamber, UAE · Abu Dhabi, UAE

What they do

IBPG is a membership association in Abu Dhabi, set up in 1991, for Indian business owners and professionals working in the UAE. Members pay a subscription and in return get networking events, a member directory, introductions, and information on tenders and on doing business between India and the UAE; it is one of the business groups registered under the Abu Dhabi Chamber of Commerce.

Inside, day to day. It is run by an elected volunteer committee of members who all have full-time jobs elsewhere, backed by a very small secretariat, so the actual daily work is organising events, chasing speakers and sponsors, keeping the membership list and renewals current, answering member enquiries by phone and email, and maintaining the website and directory.

Nobody has written a use case for this one yet.

Indic EMS

Electronics contract manufacturing · Bengaluru, India (second plant in Barcelona, Spain)

What they do

Indic EMS builds electronic circuit boards and finished electronic products for other companies: the customer designs the product, Indic buys the components and manufactures, assembles and tests it. Its buyers are automotive, telecom, medical, industrial, IoT and renewable-energy companies in India and abroad, and most of what it makes is exported.

Inside, day to day. Production operators, around 80 percent of them women, spend the day on surface-mount assembly lines placing and soldering components onto boards; engineers write test programs, quality staff inspect boards for defects, and planners and buyers juggle hundreds of component line items per order across many suppliers, largely on spreadsheets and email because every customer order is a different small batch rather than one long repeat run.

The customer test report drafts itself, in whichever format that client uses

Today a test engineer runs every device through the site's MDA, ICT, ATE and Functional Test rigs — scripted in LabVIEW and Python — then hand-writes the pass/fail numbers into whichever of Indic's 30+ active clients' own report template applies, redone from scratch every time because each client's format is different.

With AI he builds an agent that reads the raw rig log and drops the same numbers straight into whichever client's template is selected, and he edits the mapping himself the day a client changes its layout — one test run becomes one drafted report instead of one manual rewrite, and he spends the freed time on the borderline failures that actually need his judgement. That is time back.

Five minutes open a synthetic rig log on screen, labelled as such, run it against one client's report template and read out the populated report, then live in front of the room switch the target to a different client's template that adds a required "operator ID" field and re-run the same log, so the room watches the report reshape around the new format instead of repeating the last output.

@sahidmodi45

The batch write-up cites the standard it actually failed

Today a quality engineer inspects and audits each batch against the certified standards Indic holds — IPC-A-610J, ISO 9001:2015, IATF 16949:2016, ANSI/ESD S20.20-2020 and IPC/WHMA-A-620 among them — and when a batch fails, writes the root-cause analysis and corrective action up by hand, batch by batch, across a line assembling more than 10.5 million PCBAs a year.

With AI she builds an agent that reads the batch's defect data and drafts the root-cause write-up against the specific standard clause it actually failed, and she corrects the causal chain herself whenever the failure pattern is new to it — the write-up stops eating into her audit time, so more batches actually get walked and audited in the same day. That is productivity.

Five minutes open a synthetic batch defect log on screen, labelled as such, run the agent and read out the drafted root-cause note and the standard clause it cites, then live change the defect type in the log from a solder bridge to a cold joint and re-run it, so the room sees the cited clause and the root-cause hypothesis both change with the data.

@sahidmodi45

The proposal's technical scope reshapes itself when the requirement changes mid-call

Today a presales executive confers with a prospective customer and Indic's own engineers on requirements, then writes the resulting technical proposal from a blank page — for a company already running 14+ programmes per existing client, no two conversations turn up the same requirement set, so no proposal starts from the last one.

With AI he builds an agent that takes his call notes and drafts the proposal's technical-scope section against Indic's own service lines — New Product Introduction, Product Engineering, Testing and Quality, Custom Box-Build, Global Supply Chain — and he edits the wording himself as the scope shifts mid-conversation. A scope draft ready before he is off the call means more prospective engagements answered in the same week. That is revenue.

Five minutes open a synthetic requirements brief on screen, labelled as such, run the agent and read out the drafted scope section, then live — as if a customer just said so on the call — strike New Product Introduction from the brief and add box-build only, re-run it, and show the scope section drop NPI and rewrite itself around box-build alone.

@sahidmodi45

The manufacturing paperwork updates itself when the BOM does

Today a designer takes the finished board file out of Altium Designer, Eagle CAD or OrCAD and writes up the accompanying manufacturing documentation for the floor by hand — a fresh set for every product that goes through Indic's New Product Introduction line, because each one is a different customer's design.

With AI she builds an agent that reads the exported netlist and BOM and drafts the new manufacturing documentation against her past sets, and she is the one who corrects the callouts whenever a customer's design uses an unusual component footprint — the paperwork moves from a write-up per introduction to a draft she checks and tunes, and she keeps extending the callout rules herself as new component families show up. That is skill.

Five minutes open a synthetic exported BOM on screen, labelled as such, run the agent and read out the drafted documentation and its component callouts, then live swap one resistor's value in the BOM for a different one and re-run it, so the room watches the matching callout in the documentation change to match.

@sahidmodi45

The shipment list reprioritises itself when a promise date moves

Today a logistics executive coordinates shipment and inventory flow by hand across every one of Indic's 30+ active clients, each running roughly 14 programmes, under a high-mix, many-small-batches model rather than one long repeat run — and we could not establish which system, if any, sits between the company's newly upgraded ERP and this day-to-day coordination desk, so exactly what tool this runs through today is unclear.

With AI she builds an agent that reads the day's confirmed inbound shipments and open orders and drafts a priority list of which ones are actually at risk, and she edits the escalation rules herself when a client's terms change — she spends less of the day chasing every line item and more of it on the shipments that need a person's call. That is time back.

Five minutes open a synthetic multi-supplier shipment list on screen, labelled as such, run the agent and read out the priority order, then live move one shipment's promised date forward by a week and re-run it, so the room watches that shipment jump up or drop down the priority list in front of them.

@sahidmodi45

The production schedule adjusts itself when the line count does

Today a production manager builds and adjusts the schedule and tracks inventory and efficiency by hand every time customer demand shifts across the mix of 30+ active clients running roughly 14 programmes each, now spread across five more surface-mount lines than the floor had before, ten going to fifteen.

With AI he builds an agent that reads the current order book and line status and drafts the next schedule revision, and he adjusts the resourcing rules himself whenever a new line or a new client programme comes online — the revision that used to be rebuilt from scratch becomes one he checks and releases. That is productivity.

Five minutes open a synthetic order book and line-status sheet on screen, labelled as such, run the agent and read out the drafted schedule, then live add one more rush order to the sheet and re-run it, so the room watches the schedule redistribute across the fifteen lines in front of them.

@sahidmodi45

Involve Learning Solutions

Education nonprofit · Bangalore, India

What they do

Involve trains older children in Indian government schools to teach younger ones, so a student in grade 6 or 7 runs regular learning sessions for children in grades 3 to 5. It is a nonprofit paid for by donors and foundations; its customers are really state education departments and government schools that agree to let the programme run in their classrooms.

Inside, day to day. Field staff spend their days inside government schools and district offices, training teachers to run peer-learning sessions, coaching the student peer-teachers, and filling in observation forms and attendance and assessment sheets, which a small central team in Bangalore then has to chase, clean and turn into reports for governments and funders.

The paper observation sheet becomes a clean row in the spreadsheet

Today field staff fill in observation forms and attendance and assessment sheets inside government schools, largely on paper, and the small central team in Bangalore then chases them, reads them and retypes them into spreadsheets before any report can be made. I am guessing the sheets arrive as phone photos; the card only says the central team has to chase and clean them.

With AI field staff photograph each sheet, and a tool reads the photo, fills in a row per school and per session in the central spreadsheet, and flags anything unreadable, missing or out of range for a person to check; the central team edits the form layouts it knows how to read whenever a sheet changes. The team goes from retyping to checking.

Five minutes hold up three synthetic handwritten attendance sheets for made-up schools, labelled as invented, photograph them live, run the tool and show the rows appear in the spreadsheet with one smudged number flagged, then correct that number on the paper, photograph it again and show the flag clear.

@shaoibul0926

The monthly report to the education department drafts itself

Today once the data is clean, somebody in Bangalore turns it into separate reports for each state education department and each funder, writing the same kind of summary again and again in slightly different formats.

With AI a tool reads the cleaned attendance, observation and assessment data and drafts each report in that reader's format, with the key numbers, the schools that moved and the ones that need a visit; a programme person checks the figures, adds what the numbers do not show, and sends it.

Five minutes open a synthetic dataset for ten made-up schools on screen, run the tool and read the one-page district summary out, then change one school's reading scores and run it again so the room sees that school move from the needs-a-visit list to the improved list.

@shaoibul0926

Nobody has written a use case for this one yet.

Iron Lady Foundation

Community nonprofit · Palo Alto, USA, with its fieldwork in Udaipur, Rajasthan, India

What they do

Iron Lady Foundation is a small private nonprofit started in 2023 by Rashi Mehta that funds and runs projects for women and children in poor parts of Rajasthan, India. It pays for school scholarships, school toilets and clean water, sewing equipment so women can earn an income, and free medical days; it sells nothing and is funded by the founder and donations. (Note: it is a different organisation from the Bengaluru women's-leadership training company also called Iron Lady.)

Inside, day to day. No staff beyond the founder appears anywhere public, so the work runs founder-first: pick one school or one village project at a time, arrange the vendors and local volunteers over WhatsApp and email, run the camp or the class, then photograph it, write it up and report it back to donors, which means a large share of the effort is coordination and storytelling about each individual project.

Nobody has written a use case for this one yet.

JIS

Not established · India (city not established)

What they do

Not established. What is actually on record is that JIS is an organisation in India that has its own in-house content and marketing team of digital marketers and graphic designers, and its own student body; it reached us through an intermediary and all correspondence has run through his personal email, so its registered name, website and sector have never been confirmed.

Inside, day to day. The only part that can be described honestly is the team that was met: digital marketers and graphic designers who take a brief and produce creatives, campaign copy, social posts and a weekly performance report, plus a student group large enough to fill a forty-person workshop. Anyone writing use cases for JIS should treat the organisation itself as unknown until someone confirms the legal name and website.

We are not certain we have the right organisation here. Check before you write.

Nobody has written a use case for this one yet.

Kayali

Luxury fragrance brand · Dubai, UAE (offices also in London and New York)

What they do

Kayali makes and sells perfume under its own name, founded in Dubai in 2018 by Mona Kattan and known for sweet, dessert-like scents. It sells through big retailers including Sephora, Boots, Selfridges and Harrods as well as its own website, so its buyers are both retail chains that stock it and individual shoppers who wear it.

Inside, day to day. It is a lean team of roughly 180 people where most of the day goes into brand and social content, shooting, writing and scheduling campaigns for a brand built on social media, alongside retail account managers handling Sephora and Boots orders, supply-chain planners forecasting stock for well over a thousand store doors, and finance and operations staff running an ERP and an online store that were stood up in a hurry after the company was separated out of Huda Beauty.

Nobody has written a use case for this one yet.

Lals Group

Retail conglomerate · Dubai, UAE

What they do

Lals Group is a family-owned retail company in Dubai, started in 1979, that runs more than 190 shops across the Gulf, some its own brands such as Homes r Us and some franchises of foreign brands such as Daiso, Carter's and US Polo Assn. It also owns two shopping malls, Lamcy Plaza and Arabian Centre, and the distribution and logistics operation that supplies its stores.

Inside, day to day. Several thousand people work mostly on shop floors across six countries selling and restocking, while at head office buyers and merchandisers decide what each brand orders and at what price, mall teams chase tenants and leases, warehouse teams move goods out to stores, and an HR function recruits at very high volume all year just to keep 190-plus outlets staffed.

Nobody has written a use case for this one yet.

MakerGhat

Education nonprofit, India · India (Mumbai, works across Maharashtra, Karnataka, Andhra Pradesh, Tamil Nadu)

What they do

MakerGhat is an Indian non-profit that builds cheap workshop rooms, called makerspaces, inside government schools and community centres, and trains the teachers there to run hands on building projects with students roughly aged 12 to 18. Nobody pays a fee for this: the money comes from grants and donations, so in practice its two customers are the funders who pay for the work and the state education departments that let it into their schools.

Inside, day to day. Staff spend their days training teachers in batches, travelling out to hundreds of school sites to check the makerspaces are actually being used, collecting weekly assessment data by hand from each site, and then turning that pile of paper and spreadsheets into progress reports for the funders who paid for it.

Nobody has written a use case for this one yet.

Mantra4Change

Education nonprofit, India · India (Bengaluru, operating in about eight states)

What they do

Mantra4Change is an Indian non-profit that works with state and district governments to improve government schools, but it does not teach children. It trains the adults in the system instead: head teachers, classroom teachers, and government education officers, paid for by philanthropic funders and company social responsibility budgets rather than by the schools.

Inside, day to day. Roughly 150 to 175 people, most of them field staff scattered across eight states, spend their days driving to schools and district education offices, running teacher discussion groups and workshops, writing up visit notes and school progress data, while a head office research and design team turns all of that into programme designs and reports back to funders.

Nobody has written a use case for this one yet.

Mashmupa International Media

Photo and video production · United Arab Emirates (Dubai Studio City)

What they do

Mashmupa International Media is a Dubai photography and video production company that trades publicly under the Vikram Kacher Photography name. Brands, hotels, car companies, sports events and private clients hire it to shoot stills and video, including corporate portraits, product and car shoots, events, drone and aerial work, and fitness, and it hands back the finished edited files. Named clients on the public record include Armani, Nike, Mercedes Benz, Puma and Novotel.

Inside, day to day. The work comes in bursts around each booking: planning and scouting a shoot, then a day or two on location or in their own Dubai Studio City studio with lights and cameras, then much longer sitting at screens throwing away thousands of frames, retouching the ones that survive, cutting video, and going back and forth with the client until a selection is approved.

The edit starts from a shortlist, not the whole shoot

Today a shoot at the Dubai Studio City studio — whatever the genre, product through e-commerce — ends and the studio moves straight into post-production, per its own listed services. No separate editor or retoucher role shows up anywhere in the record, so, and this is a guess rather than a confirmed fact, it looks like the same small team that shot the job is the one sorting and editing every frame from it afterwards.

With AI the photographer builds a first-pass sorting tool that groups a shoot's frames by focus and exposure and tags them by shoot type, so the edit starts from a shortlist instead of the full set, and he keeps adjusting the sorting rules himself as his own eye for a keeper changes. The post-production stage shrinks and what's freed up is time for the actual retouch and the next booking — that is time back.

Five minutes open a labelled synthetic sample batch of shoot photos, run the sorting tool, and watch it split them into a keep pile and a reject pile tagged by shoot type. Then add one clearly blurred synthetic photo live, in front of the room, and run it again so everyone watches the tool catch it and drop it into the reject pile on its own.

@sahidmodi45

A new page on the site stops starting from an old one

Today Siddharth Kacher looks after the studio's graphics and website work himself, on top of marketing, production, printing, social media, branding, event consultancy, voiceovers and translation, per his own profile. The site runs on Wix, and — a guess, read off the trail it leaves rather than anything Mashmupa has said about how it works — building a new page today looks like duplicating an old one: the live site still carries a page called 'copy-of-vision-mission,' and a second one, 'copy-of-about-us-2,' now returns a dead link.

With AI he builds an agent that takes his rough notes for a new page or a new client testimonial and drafts it straight into the site's existing tone and layout, and it checks the site's own pages for orphaned 'copy-of' slugs and dead links before he publishes, so pages stop getting cloned from old ones and stop quietly going dead. He keeps tuning the drafting templates himself as the site grows — that is productivity, the same one person covering more of the site without more hand-time on it.

Five minutes open the studio's live Testimonials page next to a rough, made-up client quote, run the agent, and read out the new testimonial block it drafts in the page's existing style. Then edit that quote live — swap the client's name and the line they supposedly said — and run it again so the room sees the draft change with it, not a fixed template.

@sahidmodi45

McKinsey

Management consulting · Global, headquartered in New York, with offices in most major cities

What they do

McKinsey is a management consulting firm that big companies and governments hire when they need help deciding what to do, for example cutting costs, entering a new market, reorganising, or fixing how something is run. It sells the time and analysis of its consultants by the project, usually for a few months at a stretch, and the client pays a fee for the advice and the recommendations.

Inside, day to day. Consultants work in small teams of about three to six on one client at a time, spending their days interviewing that client's executives and staff, pulling their data apart in spreadsheets, comparing them against competitors, and turning it all into slide decks presented to the client's leadership, while the senior partners above them spend their days on client relationships and winning the next project.

Nobody has written a use case for this one yet.

MCR Consulting

Retail advisory, small firm · United Arab Emirates (Dubai), not fully established

What they do

MCR Consulting is a small independent consultancy associated with Eric Macaire, who is based in Dubai and has around twenty years of background in luxury and fashion retail, specifically merchandising, buying and retail transformation. Its own website at mcrconsulting.net currently shows only a Launching Soon page, and no public record confirms its services, client list, team size or legal registration, so what it sells and to whom cannot be stated with confidence. What is visible is senior advisory work in and around fashion, luxury and retail businesses.

Inside, day to day. Almost certainly a one person or very small operation rather than a staffed office, where the day is meetings, calls, market watching and introductions across a long standing retail network, but treat this sentence as an assumption drawn from the founder's profile and not as an established fact.

We are not certain we have the right organisation here. Check before you write.

Nobody has written a use case for this one yet.

Motivation for Excellence

Grantmaking foundation · India (Mumbai), with an affiliated US foundation in Greenwich, Connecticut

What they do

Motivation for Excellence, whose legal name is The RG Manudhane Foundation for Excellence, is an Indian charitable foundation that gives money to other non-profits instead of running programmes itself. It funds roughly twenty to thirty partner organisations across healthcare, education and training, livelihoods, and what it calls nurturing excellence, with a stated focus on rural and tribal communities, and it was set up as a family legacy rather than as a fundraising charity.

Inside, day to day. A small board of trustees, made up of the founding family plus people from finance and public health, several of them living in the United States and doing this alongside full time careers, spends its time reading funding proposals from non-profits, meeting the organisations it backs, and reading the progress reports those partners send back in order to decide whether to keep funding them.

Nobody has written a use case for this one yet.

MyChips

Mobile advertising technology · Italy (Imola), acquired in May 2026 by Mistplay, based in Canada

What they do

MyChips, the product behind the maf.ad brand and originally built by an Italian company called MyAppFree, sells an offerwall. That is the screen inside a mobile game or app where a player can try other apps or watch adverts in exchange for in game rewards, and it has two sets of customers at once: game and app studios that install it to earn money from players who would never pay cash, and advertisers who pay to put their app in front of those players.

Inside, day to day. It is an advertising business, so people there spend their days on integrations and account management: helping a game studio plug the offerwall in through a software kit or an embedded frame, negotiating and pricing advertiser campaigns, and watching dashboards of revenue per thousand impressions, fill rates and country level performance to decide which offers to show to which players.

Interactive game engine runs playable ads without manual rebuilding

Today an integration engineer tests mobile reward offerwalls and ad units by manually bundling APKs, deploying test builds to devices, and waiting for compile cycles to verify that game events and rewards trigger properly.

With AI a lightweight WebAssembly runner compiles the Python game loop directly to browser canvas, allowing the engineer to test game mechanics and offerwall triggers instantly in a browser tab without touching native build toolchains.

Five minutes open the live web build at Play Demo, run the game loop in real time, then toggle a reward parameter live to watch the offerwall event dispatch on screen.

@saiqulmodi

OpticSurg Inc

Healthcare AR software · United States (New York)

What they do

OpticSurg makes software for smart glasses that lets a nurse or doctor standing next to a patient show exactly what they are seeing, live, to a specialist who is somewhere else entirely. Hospitals, surgery centres, urgent care clinics, nursing homes and home-care companies buy it so one expert can be useful in several places in a day without getting on a plane.

Inside, day to day. It is a very small company (roughly 2 to 10 people) founded in 2017 by a surgeon who is also a biomedical engineer, so the days split between engineers writing the video and augmented-reality software, clinical staff testing it on real wards and grinding through the medical-device regulatory paperwork that any hospital product needs, and a couple of people running demos for hospital buyers and answering their procurement and security questionnaires.

The hospital security questionnaire stops starting from zero

Today — and we are inferring this from his outside IT-security consulting background rather than a duty OpticSurg has put in his own words — Michael Whitcomb, who leads Security & Compliance Strategy and separately runs the firm Cenova Cyber, answers the data-security and HIPAA questions each new hospital raises before it will clear Vision Beyond for its network. It is IT and data-security paperwork, not FDA device-regulatory paperwork, and every hospital's form asks the same handful of questions in a different order.

With AI he keeps his answers to the standard fields — encryption, access controls, breach notification, HIPAA business-associate terms — on file, and an assistant he set up drafts the next hospital's version from them, filling in only what that form asks and flagging anything it has not seen before for him to write himself. One more hospital's review gets turned around without pushing the next one back — that is time back.

Five minutes open two published hospital vendor-security-questionnaire templates side by side, since each hospital phrases the same handful of questions differently. Feed the assistant a set of made-up, clearly-labelled test answers for OpticSurg's encryption and HIPAA practices and run it against the first form, reading the drafted answers aloud. Then, live, change one test answer — from "encrypted at rest" to "encrypted at rest and in transit" — and run it against the second hospital's form so the audience watches both drafts pick up the new wording.

@sahidmodi45

Oxford

University business school · United Kingdom (Oxford)

What they do

This is almost certainly not a real organisation on this board. The domain oxford.edu is held by the University of Oxford, and the address it actually appears in (said.oxford.edu) is the mail domain of Saïd Business School, Oxford's business school, which teaches MBA and executive courses and issues lifelong alumni email addresses to its graduates.

Inside, day to day. Not established as a counterparty. The record exists because a person's personal alumni email address was turned into a company record automatically, so there is no organisation here to invent a use case for. The card should be checked against the person it came from before anybody spends time on it; if you want a real Oxford use case you would be inventing one for a university (admissions, teaching, research administration), not for anyone we are talking to.

We are not certain we have the right organisation here. Check before you write.

Nobody has written a use case for this one yet.

Partners in Prime

Community events, Dubai · Dubai, UAE

What they do

A small Dubai community run by two women who organise paid workshops and talks for parents, mostly mothers, and their teenage children, on subjects like using AI in everyday life and work. They sell tickets to each session and find their audience entirely through their Instagram page and WhatsApp groups, not through a website or advertising.

Inside, day to day. The two founders spend their days inside WhatsApp groups and Instagram messages recruiting people one by one, booking venues, agreeing the running order with whoever is teaching the session, making the flyers and posts by hand for every single event, and chasing ticket payments, no-shows and refunds afterwards; everything lives in Gmail, Instagram and WhatsApp and nothing is automated.

One event brief becomes the flyer, the post and the reminders

Today for every workshop the two founders write the Instagram post, make the flyer text, write the WhatsApp group announcement and then the reminder messages by hand, starting from a blank page each time.

With AI they fill in one short brief — topic, speaker, venue, date, time, who it is for — and a tool drafts the Instagram caption, the flyer wording, the WhatsApp announcement and a day-before and hour-before reminder, all in their own tone learned from their past posts. They edit and post it themselves; nothing is sent automatically.

Five minutes open a synthetic brief for a made-up session called "AI for busy mums", labelled as invented, run it and read the caption and the WhatsApp message out, then change the venue and the time in the brief and run it again so every piece updates together.

@shaoibul0926

The paid, unpaid and no-show list keeps itself

Today after each session the founders scroll back through WhatsApp chats and Gmail to work out who paid, who promised to pay, who did not turn up and who asked for a refund, then chase each one individually. I am guessing at exactly how payments come in; the card only says chasing them is manual.

With AI a tool reads the sign-up list and the payment confirmations, builds one table of who is paid, unpaid or asked for a refund, and drafts a polite chase message for each unpaid person. The founders check the table and send the messages themselves; refunds stay their decision.

Five minutes open a synthetic sign-up list of ten names and six synthetic payment confirmations on screen, run the tool so it shows four unpaid people with a drafted message each, then add one more payment confirmation and run it again so that name moves to paid.

@shaoibul0926

Nobody has written a use case for this one yet.

Pataka Group

Indian family conglomerate · Kolkata, India

What they do

A family-owned Indian group started in 1952 that manufactures everyday goods across five businesses: hand-rolled biris (a cheap Indian cigarette, where they are the country's largest maker at roughly 110 million sticks a day across 500+ brands), packaged tea, biscuits, silk yarn and healthcare, plus a property arm. Their buyers are Indian wholesalers and small shops reached through a distribution network of more than 150,000 outlets, and for the silk arm (East End Silks, in Malda and Kolkata) textile mills in India and abroad that buy spun silk yarn and silk top sliver.

Inside, day to day. This is a factory-and-distribution business run on spreadsheets, paper and phone calls: shop-floor supervisors tracking daily output across hundreds of biri brands and silk yarn lots, purchasing managers buying tobacco leaf, tea and raw silk at prices that move constantly and decide whether a month is profitable, quality staff logging test results by hand, and area sales officers ringing and visiting distributors every day to keep stock moving and money coming back.

Nobody has written a use case for this one yet.

PayrLabs

Software services, India · Mumbai, India

What they do

PayrLabs Softwares Private Limited is a Mumbai software company that builds and runs systems for other organisations in three separate areas: payment software for banks, fintech firms and payment processors; a healthcare platform (HealthNEXAA) for handling diagnostics and coordinating a patient's care across different providers; and an online learning product (NEXAA EDGE) that teaches job skills and career guidance to young people. Their customers are banks and payment companies, healthcare providers, and the colleges, employers and youth programmes that want people trained.

Inside, day to day. Most of the week is engineers and project managers building, testing and babysitting software that runs inside a client bank or hospital, which means a lot of requirement documents, integration work against someone else's old systems, and support tickets when something breaks; alongside that a smaller training side writes course material and runs batches of young learners through sessions, which involves recruiting each batch, tracking who turned up and who finished, and reporting outcomes back to whoever paid.

Every learner leaves with a CV scored against a real job

Today on the NEXAA EDGE training side, a trainer reads each young learner's CV one by one and tells them what to fix before they apply for jobs, and with a full batch that feedback is either rushed or skipped. I am guessing that CV help is part of the career guidance; the card only says the product teaches job skills and career guidance.

With AI a tool reads the learner's CV next to a job description they choose, scores how well it matches the way an applicant-tracking system would, lists the missing skills and weak lines, and suggests rewrites in the learner's own words; the trainer reviews the suggestions in the session and adjusts the scoring rules for each new batch's target jobs. The trainer's coaching stays the main thing; the tool clears the first pass.

Five minutes open one synthetic CV and one synthetic entry-level job post on screen, labelled as invented, run the tool and read the score and the three biggest gaps out, then add one missing skill line to the CV live and run it again so the room watches the score go up and that gap disappear.

@shaoibul0926

The batch report to the funder writes itself from the attendance sheet

Today at the end of each training batch someone combines the attendance registers, completion records and session notes by hand into an outcomes report for whoever paid for the batch, one batch at a time.

With AI a tool reads the batch's attendance and completion sheets, works out who attended, who finished and who dropped off and when, and drafts the outcomes report in the funder's format with the numbers already filled in; the programme lead checks every figure and adds the stories behind them before it goes out.

Five minutes open a synthetic attendance sheet for a made-up batch of twenty learners on screen, run it and read the drafted summary out, then mark two more learners as completed and run it again so the completion figure and the report paragraph both change.

@shaoibul0926

Nobody has written a use case for this one yet.

Quantum Partners

Corporate finance advisory · Dubai, UAE (DIFC)

What they do

Quantum Partners (DIFC) Limited is a small advisory firm in Dubai's financial district that helps business owners sell a company or part of one, buy other companies, and raise money from investors or lenders. Its clients are company owners and companies rather than the public, and it is licensed and inspected by the DIFC's financial regulator, the DFSA (reference F010044, licensed September 2024).

Inside, day to day. A handful of senior dealmakers spend their days reading company accounts and building valuation spreadsheets, writing the long information document that describes a business for sale, working the phone and their own address book to find the right buyer or investor for each mandate, sitting in negotiation meetings, and keeping a compliance and know-your-client file on every single client because the regulator can ask to see it.

Nobody has written a use case for this one yet.

Ramesh Family Office

Family office, philanthropy · Chennai, India

What they do

The private office of Ramesh Mangaleswaran, a retired senior partner of the consulting firm McKinsey & Company who founded its Chennai and Bangalore offices, and his wife Meenakshi Ramesh. It looks after the family's own money and, mainly, their giving into Indian education, skills training and disability causes; it has no outside clients and sells nothing.

Inside, day to day. A tiny team, effectively the family plus a small amount of support: reading funding proposals and progress reports from the nonprofits they back, sitting on charity boards such as Pratham Education Foundation and Bala Mandir Kamaraj Trust and preparing for those meetings, meeting nonprofit founders to pressure-test their plans, and writing up analyses and memos about where money should go, which is essentially the same consulting work done for charities instead of corporations. Note: the domain given, therameshfamilyoffice.com, does not resolve; the working one is therameshfamily.com, a private page.

Nobody has written a use case for this one yet.

Rostrum Education

University admissions consultancy · London, UK (offices in Dubai, New Delhi and Hong Kong)

What they do

Rostrum Education sells help getting into university: one-to-one admissions counselling for Ivy League, Oxbridge, boarding schools and medical or law school, plus subject tutoring for IB, A-Level, AP and IGCSE and prep for tests like the SAT, GMAT and UCAT. The customers are fee-paying families, mostly in the UAE, India, the UK and Hong Kong, who buy packages of hours with a counsellor and a tutor for a teenager applying abroad.

Inside, day to day. Counsellors sit on video calls with a student and their parents working through the same personal essay draft after draft, tutors teach scheduled lessons across four time zones, and a coordination team keeps track of which student is applying where, which deadline is next and which tutor is free at 7pm Dubai time.

The monthly report and college list stop being retyped from scratch

Today once a month, for every student on her list, a Counsellor writes up a progress report and separately compiles a spreadsheet of that student's target-college data by hand, working from her own running notes and whatever the family has added to their college list since the last round — one review names both pieces directly: "detailed monthly reports tracking my progress... helping us compile spreadsheets full of data about my colleges."

With AI she builds a tool that reads her running notes and the student's current target-college list and drafts the monthly report and the spreadsheet update from them. She still reads every line and edits the wording herself, and she is the one who re-teaches it when a family adds or drops a school. The monthly round stops being an evening spent retyping the same categories for every student and becomes a read-and-send — that is time back.

Five minutes open a synthetic student file — a page of running notes and a target-college list of four public universities, so nothing private is on screen — run the tool live and read out the monthly report and spreadsheet it drafts. Then add a fifth university to the target list on screen and run it again, so the audience watches that school's data appear in the new spreadsheet rather than a replay of what was already there.

@sahidmodi45

Saint-Gobain Nordic and Baltic

Building materials, manufacturing and distribution · Nordic and Baltic region (Norway, Denmark, Sweden, Finland, Poland)

What they do

Saint-Gobain Nordic and Baltic is the regional arm of Saint-Gobain, a very large French company that manufactures building materials such as plasterboard, insulation, flat glass and plumbing products. In this region it also owns the builders' merchant chains that sell those materials, around 400 branches trading as Dahl, Brodrene Dahl, Optimera and Tadmar, so the buyers are construction firms, plumbers, installers and contractors picking up materials for a specific job.

Inside, day to day. Roughly 14,000 people: factory shifts making insulation and board, branch staff behind trade counters serving a contractor who walks in with a materials list, warehouse and delivery crews moving stock out of six central warehouses to 400 stores, and technical advisors answering a builder's question about which wall system to use where.

Nobody has written a use case for this one yet.

ShikshaLokam

Education nonprofit · Bengaluru, India

What they do

ShikshaLokam is a nonprofit, funded by the Shibulal family, that tries to improve India's government schools by supporting the people who run them rather than by running schools itself. It builds free open-source software and training programmes for head teachers and district and block education officers, and it convenes a movement called Shikshagraha aimed at reaching a million public schools; nobody pays for it, the money comes from philanthropy.

Inside, day to day. Programme staff spend their days on workshops and mentoring calls with head teachers and district officials, a product and engineering team builds and maintains the open-source platforms those officials use to observe schools and log improvements, and a lot of hours go into writing reports, newsletters, case stories and funder updates about what changed in which district.

Nobody has written a use case for this one yet.

Shiksharth

Education nonprofit, tribal India · Sukma, Bastar, Chhattisgarh, India

What they do

Shiksharth is a small trust, started in 2015, that works on schooling for tribal children in Sukma in south Bastar, a remote district shaped by decades of armed conflict. It works alongside the district government inside existing state schools on reading and numeracy, safe spaces and locally relevant lessons, reaching a reported 480 schools and 20,000-plus children; there are no paying customers, it is funded by philanthropies such as Wipro Foundation.

Inside, day to day. Around fifty-odd educators travel out to scattered village schools, run classroom sessions and teacher training in local tribal languages, and record attendance and reading levels largely on paper; a small product team designs the learning materials and kits, and a handful of operations people handle logistics, finances and the reports funders ask for.

Nobody has written a use case for this one yet.

Solinas Integrity

Deep tech, water robotics · Chennai, India

What they do

Solinas Integrity builds robots that go inside underground pipes and sewers so that people do not have to. It sells a camera-crawler robot for inspecting water and sewer lines, a robot for cleaning septic tanks and manholes, and a software platform that turns the footage into a condition report, and the buyers are city water boards and municipal corporations across India plus industrial customers like Suez, Veolia and vehicle plants.

Inside, day to day. Field crews drive to a city, open a manhole, lower a robot in and spend hours driving it through pipe; back in Chennai engineers watch that footage to mark every crack, blockage and joint failure and write it up as a report for the municipality, while robotics and software engineers build the next version and a business development team writes government tender bids.

Nobody has written a use case for this one yet.

Tan90

STEAM education, schools · Bengaluru, India

What they do

Tan90 (legally Plumeria Learning Solutions, trading at tan-ninety.com, and not to be confused with the separate cold-chain company Tan90 Thermal) sells hands-on science and maths learning to schools. It sets up physical lab spaces called Natkhat STEAM Labs, supplies experiment kits and a structured science-and-maths programme, trains the schools' own teachers to run them, and also sells DIY kits direct to families; the paying customers are private schools, with a reported 264 schools and 293 labs across 19 states.

Inside, day to day. A small team of about forty: people writing lesson plans and designing experiments, trainers travelling out to schools to install a lab and run teacher workshops, and an operations team packing and shipping physical kits and keeping track of which school has which lab and what it needs restocked.

The cycle report stops waiting on the crunch

Today at the end of every programme cycle, Tan90's Monitoring & Evaluation Lead runs a survey of the schools with an installed Natkhat STEAM Lab, then crunches the responses into the numbers and charts the effectiveness report needs by hand before the cycle can be called finished.

With AI she builds a tool that reads the raw survey exports and drafts the report's tables and charts each cycle, and she is the one who checks the numbers against what she expects and retunes the survey questions for next time. The report stops waiting on the manual crunch to close out the cycle — that is time back.

Five minutes load a spreadsheet of invented survey responses, clearly labelled synthetic, for a made-up cohort of schools, run the tool and watch the effectiveness table and chart appear, then edit one school's response score live on screen and run it again so they watch that bar move.

@sahidmodi45

One milestone post drafts the next

Today for every new milestone — a lab install, a round of SPARC, the Summer Camp — Tan90's Communication Associate writes fresh social copy and designs the report visuals from scratch, most of it repeating the same structure the last milestone used.

With AI he builds a drafting tool, trained on Tan90's own past milestone posts, that produces the first draft of the copy and visuals from the new photos and numbers, and he is the one who edits the line that is actually new and publishes it. More milestones covered without a backlog building up behind them — that is productivity.

Five minutes open the real SPARC 2025 press release and the site's live Summer Camp 2025 page as reference, run the drafter against invented placeholder numbers for a synthetic SPARC 2026 round labelled invented on screen, read the drafted post out, then change one number in the placeholder data and run it again so they watch the copy and the stat update together.

@sahidmodi45

The next school's onboarding pack stops starting from zero

Today for every school that signs on for a Natkhat STEAM Lab, Tan90's Program Lead builds that school's training and resource pack and personally guides its teachers through the same onboarding and mentorship steps, largely rebuilding the walkthrough from the last school each time.

With AI he builds a generator, trained on the packs and mentorship notes from schools already onboarded, that drafts the next school's training pack and follow-up schedule, and he is the one who adjusts it to that school's own timetable and teachers before he runs the session himself. He is the one learning to retune it as Tan90 takes on new kinds of schools — that is skill.

Five minutes open a synthetic training pack for a placeholder school labelled invented on screen, run the generator against that school's synthetic timetable and teacher list, read the drafted pack out, then move one training session to a different day on screen and run it again so they watch the schedule paragraph change to match.

@sahidmodi45

Tier Two Capital

Venture studio, investment advisory · Patna, Bihar, India

What they do

Tier Two Capital is a small venture studio and investment advisory firm whose stated aim is to make businesses in Bihar investable. Rather than writing cheques at scale it works hands-on with a small number of local companies to get their numbers, systems and story into a state where outside investors and micro-VC funds will back them, and it advises those investors on deals in India's smaller cities.

Inside, day to day. A handful of partners plus analysts and interns embed inside a few portfolio companies at a time, cleaning up messy books and monthly management reports, building financial models, writing investment memos and diligence notes, and spending a lot of hours on calls with founders and with prospective investors including the Bihari diaspora.

Nobody has written a use case for this one yet.

Tip Tales

Animation and content studio · Singapore and Mumbai, India

What they do

Tip Tales is a small studio that makes animated films, explainer videos, brand films, websites and character designs for companies and organisations that need to explain or launch something. It also owns its own children's cartoon series, Piku N Tuki, which runs on YouTube across six languages with more than 120 episodes.

Inside, day to day. A lean in-house core of animators, editors and designers works brief by brief, reading a client brief, drawing and animating scenes, cutting versions, sending them out for review, taking change notes, re-cutting and packaging the final files for delivery, and then re-versioning the same episode into six languages, pulling in freelance sound, production and tech specialists per project.

Nobody has written a use case for this one yet.

Traveazy Group

Online travel, pilgrimage · Dubai, UAE

What they do

Traveazy runs online travel booking platforms. Its main brand, Umrahme, sells Umrah pilgrimage trips to Makkah and Madinah with the visa, flights, hotels, transport and Haramain train tickets in one booking, and its other brand, Holidayme, sells holiday packages; the buyers are individual travellers and roughly ten thousand travel agents who resell the trips.

Inside, day to day. Contracting staff negotiate and load hotel rooms and transport in Saudi Arabia into the system, a technology team builds and runs the booking site and app, visa and operations staff push each pilgrim's paperwork through Saudi government systems, support agents across the Dubai, Pune, Jeddah and Cairo offices handle changes, cancellations and problems while people are mid-trip, and account managers onboard and chase the agent network.

Nobody has written a use case for this one yet.

TrustIn

Legal tech, workplace compliance · Bengaluru, India

What they do

TrustIn sells software and services that help employers deal with workplace harassment and misconduct complaints under India's POSH law. Companies and nonprofits buy it to write their policy, train their staff, take complaints confidentially and run each case through to a decision.

Inside, day to day. The days go on casework and compliance paperwork: writing and updating a policy for each client, running training sessions including in Kannada, staffing a confidential reporting channel and a Telegram bot so an employee can report something without being identified, walking an employer's internal committee step by step through the legal process of judging a complaint, and filing the reports employers are legally required to submit, while a small product team builds the case-management tool the rest of them use.

The case-note draft writes itself before she edits it

Today — and this is inferred from the blog's own byline field, which could default to an admin account rather than confirm who actually drafted each piece — TrustIn's Founder and CEO, Meghana Srinivas, is credited on all six case-note and compliance-explainer posts published between March and September 2026, each one built around reading a full court judgment, a Bombay High Court ruling on IC jurisdiction, a Supreme Court case, and writing up what it means for employers, roughly once a month.

With AI she builds an agent that reads the text of a newly delivered judgment and drafts a first-pass case note in the structure of her last six posts, then she edits every line and decides what the ruling actually means before it goes out under her own name — the research-and-draft afternoon becomes a twenty-minute edit, and that is time back.

Five minutes open a real, public court judgment of the kind the blog already cites, run the agent to produce a draft case note, then swap in a second public judgment live and run it again, so the audience watches the draft change to match the new parties, court and ruling instead of repeating the same template.

@sahidmodi45

Tvarita Collective

Arts and culture · Bengaluru, India

What they do

Tvarita finds traditional Indian folk and classical artists and puts them into paid work. It sells live performances, storytelling shows, art workshops and handmade artist products to schools, colleges, companies, festivals and households, and pays the artists out of what it earns.

Inside, day to day. Two co-founders and a small team spend their days tracking down and verifying artists across dozens of regional art forms, matching the right troupe to an enquiry from a school or a company, quoting and scheduling the event, arranging travel and logistics for performers coming in from villages, documenting each artist's story and tradition in writing and video, and then running the event on the day.

An enquiry comes in and the right troupe is already suggested

Today when a school or company asks for a performance or workshop, a co-founder reads the enquiry, thinks through which of the artists across dozens of art forms fits the audience, the date and the city, checks who is free, and then writes back a proposal and a schedule by hand.

With AI a tool reads the enquiry against Tvarita's own artist list — art form, language, group size, where they travel from, when they are free — suggests the three best matches with a reason for each, and drafts the reply and the event schedule; the co-founder picks the troupe and edits the reply. I am guessing at how the artist list is kept today; the card only says matching is done by the team.

Five minutes open a synthetic artist list of twelve made-up troupes and one synthetic enquiry from a school on screen, labelled as invented, run it and read the three suggestions out, then change the enquiry from a school assembly to a corporate evening event and run it again so the room sees different troupes rise to the top.

@shaoibul0926

An artist's recorded story becomes a written profile the same day

Today the team documents each artist's story and tradition in writing and video, which means sitting with hours of recorded conversation, often in a regional language, and writing the profile up by hand afterwards.

With AI a tool transcribes and translates the recorded interview, then drafts a profile — who the artist is, the art form, where it comes from, what they perform — using only what was said in the recording, with the quotes marked; the team checks it with the artist before it is published anywhere.

Five minutes play a short synthetic recorded interview made for the demo, labelled as invented, run the tool and read the drafted profile out, then play one more line where the speaker corrects the name of their village and run it again so the profile updates to match.

@shaoibul0926

Nobody has written a use case for this one yet.

Vidhya Vidhai

Education nonprofit · Chennai, Tamil Nadu, India

What they do

Vidhya Vidhai is a nonprofit that trains head teachers and teachers in government and low-income schools across Tamil Nadu and Puducherry to run their schools better. The schools do not pay: funders such as Wipro Foundation pay for the programmes, and one idea they built, a student house system called Magizh Mutram, was picked up by the Tamil Nadu education department and rolled out across government schools.

Inside, day to day. Around twenty-five to forty staff, split into district field teams (Coimbatore, Nagapattinam, Tirunelveli, Cuddalore, Sivagangai) and central teams for programme design, research, people and culture, growth and communications, spend their days travelling out to schools, running leadership workshops with head teachers, writing up visit notes and filling trackers and reports afterwards, and gathering data across hundreds of schools to show funders what actually changed.

The needs-assessment write-up drafts the two-year plan

Today — and we are guessing at the exact title here, since Vidhya Vidhai's own account only says "our team" — the Programme Design Lead pulls together five separate threads of a new school's needs assessment (classroom observations, teacher interviews, school-leader interactions, student assessments, parent interactions) by hand into a two-academic-year implementation plan, then repeats the same synthesis at baseline and again at endline for every school or cluster the programme takes on.

With AI she builds a drafter that reads the five assessment inputs and produces the implementation plan in the organisation's own structure, and reruns it on the baseline and endline notes to draft what changed; she edits it herself each time a new school's context is different. The plan and the change write-up come back as a draft to edit instead of a blank page to fill — that is time back.

Five minutes open three synthetic assessment notes on screen, clearly labelled "synthetic, for demonstration" — a classroom-observation note, a teacher-interview note, and a parent-interaction note for a made-up school — and run the drafter to produce a one-page plan. Then, live, edit one line in the teacher-interview note, changing a stated weakness from "lesson planning" to "attendance tracking," and rerun it in front of the audience so the plan's priority-actions section visibly changes to match the new note, showing it is reading the actual input, not filling a template.

@sahidmodi45

Zerodha

Retail stockbroking, fintech · Bengaluru, India

What they do

Zerodha is India's largest stockbroker. Ordinary people open an account and buy and sell shares, mutual funds and derivatives through its apps, Kite for trading, Coin for mutual funds and Console for statements, and Zerodha charges a flat fee of up to twenty rupees per order, with nothing charged on ordinary share purchases.

Inside, day to day. It serves more than ten million customers with only about a thousand staff and no sales team, no advertising and no sales targets, so most of the work is support and back office: answering an enormous daily volume of customer tickets and phone calls, checking identity documents to open new accounts, and doing the daily regulatory filings the exchanges and the regulator demand, alongside a famously small in-house engineering team (reported at around thirty-six people) that builds the trading platforms and a team that writes free investor-education material called Varsity.

The support agent starts from a drafted answer, not a blank reply

Today a support agent opens a ticket — why was this charge applied, why is my order rejected, how do I change my bank account — and searches Zerodha's own support pages and Varsity material to write the answer, many times a day, when a large share of tickets are close repeats of each other. I am guessing at the share of repeats; the card only says ticket volume is enormous.

With AI an assistant reads the new ticket, sorts it into a category, pulls the matching support-page passages and drafts an answer that quotes them, with the source linked; the agent checks it against the customer's account and sends or edits it. The support team keeps the list of source pages it is allowed to answer from, so a changed rule is one page edit away.

Five minutes open five synthetic tickets on screen, labelled as invented, run the assistant so each one shows its category, the drafted answer and the support page it came from, then edit one ticket's question from a charge query to a bank-change query and run it again so the room sees the answer and the source page both change.

@shaoibul0926

Account-opening documents are checked before a person looks

Today a back-office staff member opens each new account application and checks the identity documents by eye — is the photo readable, does the name match the form, is anything missing — before approving it.

With AI a tool does the first pass, reading each uploaded document, comparing the name and date of birth with the form and flagging blurry images, mismatches or missing pages, so the staff member spends time only on the flagged ones and still makes every approval decision.

Five minutes use three obviously fake sample documents made for the demo, never real ones, run the checker and show two passing and one flagged for a name mismatch, then correct the name on the form and run it again so the flag clears.

@shaoibul10926

Nobody has written a use case for this one yet.

Zuva Life

Midlife transitions, coaching · Bengaluru, India

What they do

Zuva Life runs programmes for people in midlife and beyond who are working out what they do next, covering longer careers, new kinds of work, money, learning and staying connected. It sells workshops, small-group cohort journeys, retreats and one-to-one sessions, mostly through employers, wellness and longevity companies, career platforms and coaches rather than straight to individuals.

Inside, day to day. A very small team around founder Vandana Srivastava, with a co-founder, an associate, a brand designer and a bench of outside coaches and advisors, spends its days designing session content and exercises, recruiting and then running each cohort, holding one-to-one conversations with participants, writing brand and social content, and pitching the programme to HR teams and wellness partners who will bring a group.

Nobody has written a use case for this one yet.

Book Demo with Sarfaraj

AI enablement training · demo & full-course classes

What they do

Stratos Tech Academy runs AI-enablement training for people who want to get hands-on with AI tools rather than just read about them: a short demo class for anyone curious, and a full course for anyone who wants to go deeper. Led by Sk Sarfaraj Ali.

Inside, day to day. Interested people reach out over chat, email or social DMs to ask about the demo class or the course, a slot gets found by going back and forth on availability, and then the session itself gets run and followed up on one at a time.

Booking a seat stops living in an email thread

Today someone interested in the demo class or the full AI-enablement course messages back and forth to find a slot that works, often losing a day or two to timezones and missed replies before anything lands on a calendar.

With AI a booking link shows real availability and confirms a 30-minute demo slot instantly, then sends the class link and prep notes automatically, no manual back-and-forth needed.

Five minutes open the booking page live, pick a real slot, book it on the spot, and show the confirmation that goes out immediately.

Sk Sarfaraj Ali