- A captive engineering center puts India engineers on your own payroll, so the hard part is never the setup. It is the running.
- Pods of six to ten engineers with one named owner beat large shared pools, because accountability survives a nine hour time gap.
- Two to four hours of genuine overlap with a US team is plenty, provided you spend it on decisions instead of status updates.
- Track deployment frequency, lead time, change failure rate, and recovery time. Never track hours logged or tickets closed.
- You can turn a vendor or ODC team you already work with into your own captive team through an EOR, without registering an entity first.
Need help with captive engineering centers in India? Talk to an expert!
Discover how Wisemonk creates impactful and reliable content.
You have board approval, the entity is registered, and the first ten offers are signed. So why do so many captive engineering centres in India stall out around month seven?
Because setting one up is a project with a finish line. Running one is a management problem that never ends, and almost nobody briefs you on the second part.
This guide is the second part. If you are still deciding whether to build one at all, start with our complete guide to captive centers in India, which covers the business case, the working models, the city hubs, and what setup costs.
Everything below assumes that decision is already made. We are going to talk about pod design, coverage windows, the four metrics that actually predict delivery, keeping your best engineers past year two, and the move most teams never realize is on the table.
What does a captive engineering center in India actually own?
It owns outcomes, not tasks. Your India engineers sit on your payroll, commit to your repos, join your standups, and carry your on-call pager. A vendor gets measured on hours delivered. A captive team gets measured on what shipped and whether it stayed up.
In practice, a working India center owns some mix of these five things:
- Product surface area: one or more services, features, or apps end to end, including their bugs at 2am.
- Platform and infrastructure: CI/CD pipelines, observability, cloud cost, and internal developer tooling.
- Sustaining engineering: the migrations, upgrades, and dependency work headquarters keeps deferring to next quarter.
- Quality engineering: test automation and release verification owned inside the pod, not handed to a separate QA group.
- On-call and incident response: real production accountability, not business-hours triage that escalates everything west.
India now hosts 2,117 global capability centers across 3,728 individual centers, employing about 2.36 million professionals and generating $98.4 billion in annual revenue in FY2026. Source: Wisemonk, India GCC landscape.
That scale is why the model works. It is also why your center competes for talent with roughly two thousand other well funded engineering organizations, most of them clustered in the same handful of GCC hub cities. Which brings us to how you structure the team.
How should you structure engineering pods in an India captive center?
Build pods of six to ten engineers, each with one named owner and one clear service boundary. Pods this size survive a nine hour time difference, because most decisions can be made inside the pod without waiting for someone in another hemisphere to wake up.
What goes into a single pod
- One engineering manager or tech lead: accountable for what ships, not a coordinator who forwards tickets westward.
- Four to seven engineers: weighted senior and mid-level, with at most one junior per pod in the first year.
- One quality engineer: embedded in the pod and reviewing the same pull requests, not pooled in a shared testing team.
- A named counterpart at headquarters: one product or engineering person who owes this pod answers within one working day.
Two structures that reliably fail
- The staffing pool: engineers reassigned weekly to whatever is urgent. Nobody owns a system, so nobody improves one, and your best people leave first.
- The mirror team: an India pod shadowing a US pod on the same codebase. It manufactures review queues and quiet resentment rather than capacity.
Structure only gets you so far, though. The pod still has to work across a time zone gap, and that is mostly a question of ritual rather than org chart. Our notes on managing offshore teams in India go deeper on the day to day habits.
How does follow-the-sun coverage work between a US team and an India center?
India Standard Time runs 9.5 to 12.5 hours ahead of the US mainland, which leaves a natural two to four hour overlap with the US East Coast during the India evening. Treat that window as your most expensive meeting room and only spend it on decisions.
- The overlap window: 8:00 to 11:00 am US Eastern lands at roughly 6:30 to 9:30 pm in India. Use it for design review and unblocking, never for status.
- Written handoffs: end-of-day notes in the pod channel so the US morning opens with context instead of a queue of questions.
- A decision latency budget: anything blocking work gets answered inside one overlap window, or it gets escalated automatically. Unblocked work is the whole game.
- A self-contained on-call rotation: India runs primary on-call for the services India owns. Splitting one service's rotation across two hemispheres is how incidents get dropped between them.
Get the rituals right and the time difference stops being a tax and starts being coverage. But you still need to know whether the center is actually performing, and this is where most reporting goes wrong.
Which delivery metrics should you hold an India engineering center to?
Hold it to the same four metrics you would hold any engineering team to: deployment frequency, lead time for changes, change failure rate, and time to restore service. These are the four measures popularized by Google's DORA research, and they work precisely because none of them can be gamed by working longer hours.
| Metric | What it actually tells you | Why it suits a distributed team |
|---|---|---|
| Deployment frequency | Whether the pod can ship without asking permission | Low frequency usually means an approval is sitting in another time zone |
| Lead time for changes | How long a commit waits before it reaches production | Exposes review and handoff delays caused by the overlap window |
| Change failure rate | Whether speed is costing you stability | Catches a pod that hits velocity targets by skipping tests |
| Time to restore service | Whether the team can genuinely operate what it owns | The clearest signal that on-call ownership is real rather than nominal |
What not to measure
Three metrics do active damage in an offshore context, because they were designed to police vendors:
- Hours logged: you stopped buying hours the moment these engineers joined your payroll. Counting them signals you still think you rent them.
- Tickets closed: rewards splitting work into small tickets and punishes anyone who fixes a root cause.
- Story points per sprint: points are a planning aid, not an output. Comparing them across two pods in two countries produces nothing useful.
Good metrics tell you whether the center works. They will not keep it staffed, and staffing is where India centers get genuinely hard. Our playbook for building an offshore team in India covers the sourcing side in more detail.
How do you keep engineers from leaving your India center?
By competing on the things a large Indian IT services employer structurally cannot offer: product ownership, direct access to the people who decide the roadmap, and a title that means something outside your company. Pay matters, but pay alone loses to the counter-offer. Our guide to managing attrition in India covers the wider picture.
Four things move retention more than a raise:
- A real career ladder: written levels with written expectations. Ambiguity about promotion is the single most common reason a strong senior engineer starts interviewing.
- Visible scope: let India engineers present to leadership and write the design docs. Invisible work is unrewardable work.
- Equity that they understand: grants only retain people who know what they hold, so explain the mechanics and the tax treatment. See equity compensation in India.
- Planning around long notice periods: Indian tech contracts commonly run 30 to 90 day notice, and 90 days is normal at senior levels. That cuts both ways, so read how notice periods work in India before you plan a backfill.
There is one more option worth knowing about, and it is the one that surprises people. You may not have to hire your captive team from scratch at all.
Can you convert an existing vendor or ODC team into your own captive center?
Often yes, and it is usually faster and less risky than hiring thirty strangers. If you already run an offshore development center in India through a vendor, those engineers already know your codebase, your product, and your quirks. Rebuilding that context from zero is the expensive part of a captive center, not the paperwork.
The move has four gates, and they should be checked in this order:
- Read your MSA first: the real constraint is almost always the non-solicitation clause in your contract with the vendor, not Indian employment law. Some clauses expire, some carry a buyout fee, some are narrower than they look.
- Separate two different legal questions: India voids post-employment non-compete clauses against employees under section 27 of the Indian Contract Act, but a non-solicit between two businesses is a separate matter and can bind you. Read our primer on non-solicitation agreements and take local advice before you approach anyone.
- Decide what employs them on day one: you do not need a registered entity to start. An India Employer of Record can become the legal employer in weeks, so the team transfers once rather than twice while your entity is still being incorporated.
- Re-paper the IP chain: under the vendor model your IP assignment ran through the vendor. New employment contracts need fresh assignment and confidentiality terms signed directly with you. See protecting IP with India developers.
Plenty of companies run the center on an EOR for the first year or two and incorporate later, once headcount justifies the overhead. The trade-offs are laid out in EOR vs entity in India, and it is a reversible decision, which is more than can be said for most of this list.
What does it cost to run a captive engineering center in India each year?
Budget $25,000 to $80,000 per engineer per year fully loaded, which lands 40% to 60% below equivalent US operating costs and 50% to 65% below at senior levels. That is the running number, and it is an estimate rather than a quote. If you need the one-time build number instead, it sits in our guide to GCC setup costs in India.
The gap between a good estimate and a bad one is rarely salary. It is the additions people forget:
- Statutory employer contributions: provident fund, insurance, gratuity accrual and state levies sit on top of gross salary. The detail is in the true cost of employment in India.
- Annual increment cycles: Indian tech salaries reset yearly and senior specialists reset faster. A three year plan built on year one salaries will be wrong by year two.
- Replacement cost: every departure carries recruitment fees, a 30 to 90 day notice overlap, and a ramp period. Model attrition as a line item, not a surprise.
- Equipment and workspace: laptops, security tooling and either desks or a home-office allowance. Our employee cost calculator will size a single hire quickly.
Even with the budget right, the first year has a recognisable shape, and knowing it in advance is most of the cure.
What usually goes wrong in the first year, and how do you avoid it?
Almost every struggling center we see has the same four problems, and none of them are about engineering ability.
- Hiring engineers before hiring a leader: a pod with no local manager reports to someone twelve hours away and quietly drifts. Hire the engineering manager first, even if it costs you six weeks against the India hiring timeline.
- Starting with low-stakes work: giving the new team only backlog cleanup to prove itself teaches everyone that India does the leftovers. Give a real service in the first quarter.
- Ignoring permanent establishment exposure: how you contract and who signs what can create a taxable presence in India. Understand permanent establishment risk in India before you scale, not after.
- Treating compliance as a formality: India's four labour codes have been in force since 21 November 2025, and state rules are still rolling out as of August 2026. Track India's new labour codes rather than assuming last year's handbook still applies.
One last distinction worth drawing. If the team you are building will invent and patent rather than build and operate, the tax and IP questions change substantially, and we cover those separately in captive R&D centers in India.
How does Wisemonk help you build and run an India engineering center?
Wisemonk is an India-native Employer of Record. We help global companies hire, pay, and manage employees in India, and we take on the compliance work sitting behind every payroll cycle so your engineering leaders can spend their time on delivery instead of statutory filings.
More than 300 global clients work with us, we manage over 2,000 employees, we process $20M+ in annual payroll, and we hold a 4.8 out of 5 rating on G2. Our EOR pricing starts from $99 per employee per month.
For teams standing up an engineering center, five things tend to matter most:
- Employment without an entity: we become the legal employer in India, so your first pod can start while incorporation is still in progress.
- Engineering recruitment: our India recruitment support sources and screens for engineering and leadership roles across the major tech hubs.
- Payroll and statutory compliance: managed India payroll with contributions, deductions and filings handled on schedule.
- Equipment and asset recovery: we procure laptops and recover assets from remote employees in India when someone exits.
- IP and contract paperwork: employment agreements with assignment and confidentiality terms that protect your intellectual property in India.
An entity you own outright: where the goal is a captive center of your own rather than an EOR arrangement, we build and operate one that is yours from incorporation.
Wisemonk Entity: an engineering center you own, operated by us
For an engineering center this matters more than it first sounds. The entity employing your engineers also holds their output, so owning it from the start removes a negotiation you would otherwise have at handover.
Standing one up yourself means 6 to 12 months, 20-plus registrations and filings, and a resident director you do not have. It runs in four stages instead:
- Build: the entity incorporated, registered and banked in weeks.
- Operate: we run compliance, payroll, people and banking, so your engineering leads are not doing statutory filings.
- Graduate: we hand over the director seat, banking and records whenever you are ready, for a one-time transition fee rather than a penalty.
- Own: you hold 100% of the equity at every stage. Unlike nominee-ownership models, nobody else ever holds your shares.
How much we run is your choice, across three levels:
- Nominee and Compliance: we hold the resident director seat and carry statutory compliance, and you run banking.
- Managed Operations: we operate the entity day to day, and you approve the large payments.
- Fully Operated: we act as your outsourced India COO and finance function.
Every level includes the same operating layer, coordinated by one accountable team rather than five vendors:
- The mandatory local seat, filled: a resident director, plus ROC, GST and TDS compliance and every periodic filing.
- The full employee lifecycle on one platform: hire, onboard, pay, manage, equip, support and offboard, run as a real product rather than a spreadsheet-and-email service.
- Everything around the team: banking within limits you set, engineering recruitment, equipment provisioning, managed office space through WeWork and partners, and inter-company MSAs with in-house counsel.
- De-risked by design: D&O and professional indemnity cover, with authority limits you define in writing.
On pricing, keep the two apart. $99 per employee per month is the Employer of Record rate, published and flat. Entity and GCC work is always a custom quote.
That quote is a one-time setup fee plus a monthly management fee combining a base with a per-employee element, where the base scales with the service level you pick.
It fits three situations in particular: market entrants appointing India teams, offshore builders scaling delivery beyond an EOR, and companies graduating off an EOR into an entity of their own.
We provide EOR services in India, and we are expanding rapidly into the US and UK markets.
What do clients say about building engineering teams with Wisemonk?
Companies from the US, UK, and Europe trust us to build their teams compliantly and fast. Here's what our clients say, and you can read more customer reviews:
I'm very happy that I discovered Wisemonk. They have been a pure pleasure to work with, and their attention to detail is impressive. They helped us understand their pricing model, find top-qualified individuals, interview them, and then onboard them. I gave them criteria for the type of people we sought, and they delivered. The individuals they were able to find have been some of the best engineers I have ever worked with. I recommend Wisemonk to anyone who is in need of staffing assistance.
- Dan Sampson, Head of Engineering at Cobu
Standing up an engineering center in India?
We will employ your first pod in weeks, without waiting on entity incorporation.
Frequently asked questions
What is the difference between a captive engineering center and a GCC?
In practice they describe the same thing. Global capability center is the current market term and covers any owned offshore unit, while captive engineering center specifies that the unit does engineering. GCC is now the more widely used label for the same operating model.
How many engineers justify a captive engineering center in India?
Around 15 to 20 engineers is where a captive model usually starts making sense, because that supports two pods plus local leadership. Below roughly 10 people, an Employer of Record arrangement is normally cheaper and faster than owning the operation outright.
Can you run a captive engineering center in India without an entity?
Yes. An Employer of Record becomes the legal employer in India, so your team can start working within weeks while incorporation runs in parallel. Many companies operate this way for a year or two, then move to their own entity once headcount justifies the overhead.
How long does it take to hire an engineering team in India?
Expect four to ten weeks per hire from open role to signed offer, then a notice period on top. Indian tech contracts commonly run 30 to 90 days notice, and 90 days is normal at senior levels, so plan start dates around that rather than the offer date.
Who owns the IP created by engineers in an India captive center?
You do, provided each employment contract contains a valid assignment clause signed directly with the employing entity. If the engineers previously worked through a vendor, the old assignment ran through that vendor, so the paperwork has to be redone when they transfer to you.
Can you move a team from an outsourcing vendor to your own captive center?
Frequently yes, but the binding constraint is the non-solicitation clause in your contract with the vendor rather than Indian employment law. Review that clause and take local legal advice before approaching anyone, since terms, expiry dates and buyout fees vary widely between agreements.
What time zone overlap do you get between an India center and a US team?
India Standard Time is 9.5 to 12.5 hours ahead of the US mainland, giving roughly two to four hours of overlap with the East Coast in the India evening. That is sufficient if the window is reserved for decisions and design review rather than status reporting.
Ready to build your India team?
Tell us who you're looking to hire. We'll walk you through exactly how the setup works for your company, your timeline, and your budget.