For
Nano GCCs
Nano GCCs of 5–50 people in Karnataka
A nano GCC is a global capability centre of 5–50 people. Karnataka's GCC Policy 2024–2029 names the category and extends incentives with no minimum threshold. Chokmah helps a nano GCC ship one agentic workflow in a quarter: a free Reality Check, a light diagnostic, then a sprint that leaves code the team owns.
- A nano GCC is a capability centre of 5–50 people, named explicitly in Karnataka's GCC Policy 2024–2029.
- The policy targets 1,000 GCCs by 2029, 350,000 jobs and US$50B in output, with no minimum threshold for nano centres.
- At this size, one shipped workflow beats three demos: a cohort is the wrong first purchase.
- The realistic first engagement is a free Reality Check, a light diagnostic, then one Workflow Sprint.
- Only about 5% of custom enterprise AI tools reach production (MIT NANDA, 2025), so proving one is the goal.
Where it hurts
No incumbent vendor and no AI function
A 5–50 person centre has nobody whose job is AI, no master-vendor already embedded, and no internal platform team. That is a disadvantage for a big rollout and an advantage for a small, focused first project. There is no politics and no legacy to unwind.
One workflow is the whole opportunity
At this size the centre does not need a transformation programme. It needs one workflow that runs faster or with fewer errors, shipped in weeks, that proves the model to the parent and the founder before more is spent.
Policy incentives are real but unclaimed
Karnataka's GCC Policy 2024–2029 extends incentives to nano GCCs with no minimum employment or investment threshold, but claiming them and building an AI capability that justifies the setup are two different problems, and the second is unaddressed.
Easy to buy the wrong thing first
With a small budget and pressure to show progress, a nano GCC is most likely to buy a cohort or a course catalogue: activity that produces certificates and no changed workflow. The first spend is the one most likely to be wasted.
The market, with sources
Karnataka's GCC Policy 2024–2029 defines a nano GCC as 5–50 employees and extends incentives with no minimum employment or investment threshold.
The policy targets 500 new capability centres (1,000 total by 2029) with 350,000 jobs and US$50B in economic output.
MIT found only about 5% of custom enterprise AI tools reach production, and 95% of pilots showed no measurable P&L return.
What a nano GCC is actually dealing with
A nano GCC (a global capability centre of 5 to 50 people) is the most realistic first sale we have, and also the one where the first decision matters most. There is no incumbent vendor to displace, no platform team, and no politics. There is also very little room to waste the opening budget.
Karnataka's GCC Policy 2024–2029 is the reason there are more of these centres every quarter. It names the nano-GCC category explicitly and, unusually, attaches incentives with no minimum employment or investment threshold. The policy gets you the centre. It does not tell you what to automate first: that is the question we answer.
The first 90 days, done deliberately
For a small centre with one workflow and no AI function, the plan is short and executable, not a strategy deck.
- 1Weeks 1–2
Name the one workflow
A short diagnostic finds the single workflow worth automating first: high-frequency, rule-dense, already producing a record you can baseline. One, not five.
- 2Weeks 3–6
Ship it in a sprint
Build the agentic solution with your own two or three people so the capability stays after we leave. You own the code, the evaluation harness and the runbook.
- 3Weeks 7–12
Prove it, then decide
Measure against the baseline agreed up front. If it worked, the parent has evidence to fund more. If it did not, you spent weeks, not a year, finding out.
What does not work here
What does not work for a nano GCC is buying a cohort first or spreading a small budget across several half-built pilots. At this size, one shipped workflow beats three demos. We will steer you away from the Capability Cohort until there is something worth scaling.
Where we would start
With a free AI Reality Check to see whether there is a real workflow to build, then a lightweight Adoption Diagnostic and a single Workflow Sprint. The support ticket deflection scenario is close to the shape a nano GCC's first project often takes.
Who this is not for
This path assumes a genuinely small, new centre. If you are a large IT services firm or one of the Fortune 500 MNC GCCs, the nano playbook does not apply and we are not the right practice for you.
Ship one workflow in your first quarter
Start with a free AI Reality Check. If there is a real workflow to build, we will scope a single sprint that leaves working code your small team owns.
How we help
- AI Reality CheckNinety minutes, on-site, where your team builds one real thing from their own work.
- Adoption DiagnosticTwo weeks to a scored report naming which three workflows to automate, which to leave alone, and why.
- Workflow SprintFour to six weeks, one named workflow, built with your people. They own the code.
Relevant scenarios
Frequently asked questions
A nano GCC is a global capability centre with 5 to 50 employees. Karnataka's GCC Policy 2024–2029 is the first Indian state policy to name the category explicitly and extend incentives to it with no minimum employment or investment threshold, which is why small, new capability centres are appearing across the state.
One workflow, not several. The right first candidate is high-frequency, rule-dense and already produces a record you can baseline: often a support, document-handling or reconciliation task. With a small budget and pressure to show progress, spreading spend across several half-built pilots is the most common and most expensive mistake.
With help, one production workflow and the capability to maintain it: built by two or three of its own people so the skill stays after the vendor leaves. What it cannot realistically do alone is stand up agent orchestration, an evaluation harness and governance from scratch while also running the business, which is where a short, focused engagement fits.
The Karnataka GCC Policy 2024–2029 incentivises setting up capability centres, including nano GCCs, but it does not answer the operational question of which workflow to automate or how to build an AI capability that justifies the centre. The policy gets you the centre; the AI work is a separate problem that Chokmah addresses.
See what a two-week diagnostic finds
We interview your people, shadow two workflows, and score which three to automate, and which to leave alone.