Colleague-taught 20-minute clinics on the AI workflows a team already uses
TL;DR Knackroom finds the people in a tech team who have quietly worked out how to use AI coding and writing tools well, and helps them run short weekly clinics for their colleagues. L&D or delivery heads at IT firms and GCCs pay per seat, because vendor courses rarely match a team's own stack.
Every large Indian IT services firm is retraining staff as generative AI spreads, according to a rating agency's December 2024 review[1]. Global capability centres, with over 1.9 million staff across 1,760 centres[2], are doing the same.
The training usually shows up as generic vendor courses. Meanwhile the useful stuff sits with a few people per team who have figured out which prompts work on their codebase, which outputs not to trust and where the tools save real time. That knowledge spreads by accident in hallway chats, stays stuck in one team and leaves when those people leave.
I found no Indian survey that measures this, so treat it as a hunch to test in interviews.
Knackroom gives each team a simple loop. Engineers nominate or volunteer a tip they use, like generating test cases for a legacy module. The platform schedules a 20-minute live clinic, hands the presenter a one-page template and records it. A week later, attendees say whether they tried the tip. Clips go into a searchable library tagged by tool, language and task.
Managers see which tips got tried, which teams teach the most and which skills nobody's sharing yet. Presenters get visible credit that can feed into appraisals.
The product stays thin on purpose - scheduling, templates, recording and follow-up nudges. The content comes from the team, and that's the whole point, since it matches their own code and clients. It is not a course catalogue and doesn't try to be one.
Engineers attend and present. The L&D head or a delivery head with an upskilling target pays. Price per active seat, say ₹200–500 a month (my assumption, no public benchmark found), well below what firms pay for course libraries. A team lead with a small tooling budget can also start a pilot without waiting for central approval.
Retraining for generative AI is now a stated priority at the largest firms[1]. Teams have the tools but very uneven skill with them. Short video recording and transcription are cheap now, so a clinic library costs very little to run.
There's no public figure for peer-learning software in India. On the adjacent side, the mid-market GCC segment alone counts 480+ centres with over 210,000 staff[2], and each one is a plausible first buyer of a few hundred seats. Size the business from pilots, not analyst reports.
One global collaborative-learning platform lets internal experts build courses and claims 2,500+ customers[3]. An Indian learning platform offers LMS and LXP tools with short 3–5 minute content and AI authoring[4]. Both are full platforms. The gap is a light layer that runs live, team-level clinics on the tools people use today.
| Difficulty | Moderate (30/100) |
| Starting capital | ₹35–70 L |
| Time to launch | 4–6 months |
The software is simple. The hard part is behaviour, which means getting busy engineers to present and keeping attendance up once the novelty wears off. You also need a way to show managers that clinics changed how people work and not just how many showed up.
No special licence here. Peer learning software has no sector permit, so you only need the baseline.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't trust AI to catch every piece of client code in a recording. A presenter or team lead should review a clip before it's shared widely.
Want to work out the AI layer for Knackroom? Talk to Yash →
I like this as a cheap, testable idea for a founder who has run engineering teams and knows how hard it is to get people to share. If you want a content business then this isn't it. Prove first that 40% of attendees try a tip within a week, across three teams.
Plan for roughly ₹35–70 L. The software is light, so most of it goes on a small product team and on pilot sales to IT firms and GCCs.
No sector licence. You need company registration, GST once over the threshold, a Shops and Establishments registration and DPDP compliance for recorded sessions.
Yes, any tech team adopting AI tools has the same uneven skills. Check local employee privacy rules on recording staff first.
The free Knackroom validation kit has the line-by-line capital breakdown, the effort scoring, the data-protection checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for engineers and L&D heads, and 5 assumptions to test. Download the kit →
Thinking about building Knackroom or something close to it? The make-or-break work is keeping engineers presenting after the first month, and I've got a few thoughts on that if you want them. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.