Startup Deep Dive : HyperVerge — how a bootstrapped IIT Madras startup reached Rs 149 crore on about $1 million raised

In December 2016, a small team of IIT Madras engineers had roughly six months of cash in the bank and a double-digit-million-dollar offer to sell the company on the table. They said no. Nearly a decade later that same company, HyperVerge, reported operating revenue of ₹149.2 crore (about $15.5 million) for the year to 31 March 2025, and says it has run identity checks on more than a billion people, all while raising close to nothing by startup standards.

The number that frames the whole story is not the revenue. It is the funding: HyperVerge has raised about $1 million in external capital, in a single seed round back in 2015, and describes itself as a bootstrapped business. Most software companies that reach nine-figure rupee revenue have burned through many multiples of that. HyperVerge is the rare Indian software-as-a-service firm that grew the opposite way, and this piece traces how it got there, what it actually sells, and where the strain is now showing in its accounts.

Quick facts

What HyperVerge does

HyperVerge sells software that lets a bank, lender, insurer or app confirm that a new customer is a real, specific person, and do it in seconds rather than days. It is business-to-business and enterprise-facing, not a consumer app.

• Core products: identity verification, Know Your Customer (KYC), Know Your Business (KYB), face-match and liveness checks, document verification, and an end-to-end onboarding workflow (company, 2025).
• Who buys it: banks, non-banking financial companies (NBFCs), digital lenders, insurers, crypto and wallet firms, plus gaming, edtech and logistics companies (YourStory, 2025; company).
• The job it does: the company states its stack reduces sign-up drop-offs and blocks first-party and third-party identity fraud while keeping customers inside KYC and anti-money-laundering rules (company, 2025).
• Scale claimed: more than one billion identities verified across 195+ countries and 200+ customers, company-stated as of 2024–2025 (company; YourStory). Earlier company material put the figure at 700 million–750 million, so the billion mark is recent.

The origin: a computer-vision lab at IIT Madras

HyperVerge did not begin as an identity company. It began as a Computer Vision Group at IIT Madras, where a set of students went deep on deep learning before deep learning was a buzzword. The company was incorporated on 30 December 2013 and incubated by the IIT Madras Incubation Cell with a first cheque of about ₹6 lakh, as YourStory reported at the time. The founding group was five strong: Kedar Kulkarni, Vignesh Krishnakumar, Kishore Natarajan, Praveen Kumar and Saivenkatesh Ashokkumar.

The founding insight was narrow and technical: the same image-recognition engine that could read a photograph could also read a face, a passport page or a utility bill, and do it at machine speed. What the team did not yet have was a paying market. Their first years were spent hunting for one, taking on hard industrial vision problems because those were the problems in front of them. According to a Sify profile, an early deep-learning project cut the cost of a rail overhead-line inspection system from crores to about ₹2 lakh, and the team also did work for MRF on tyres and ITC on biscuits. Impressive engineering; almost no revenue.

The struggle years

Between roughly 2014 and 2016 HyperVerge looked less like a company and more like a very talented lab searching for a business. Two things nearly ended it.

First, the consumer bet. The team built a photo-organisation app to show off its image engine, and later experimented with a health app that read the calorie content of food on a plate, as described in Sify’s account. Neither turned into a business. The Indian Railways project that had shown such promise stalled over contract and legal issues, so the cleverest early win produced no durable income. For a bootstrapped team, months of building product that no one paid for is the near-death experience.

Second, the money simply ran low. Even after a $1 million seed in 2015, a team without a repeatable revenue engine burns runway fast. By late 2016 the founders were, by their own telling, down to roughly half a year of cash. The pivot that saved them was unglamorous: stop trying to delight consumers, and instead solve a boring, mandatory, high-volume problem for businesses. That problem was verifying who a customer is, which every regulated financial company in India and much of Asia has to do, over and over, by law.

The turning point: saying no to a buyout

The single decision that defines HyperVerge came in December 2016. With the pivot to identity verification just beginning and only months of cash left, the founders received an acquisition offer worth, in Sify’s words, double-digit millions of dollars; secondary accounts put the range at roughly $10 million to $12 million. Selling would have handed a young team a life-changing outcome and removed all the risk. They turned it down.

The bet behind that refusal was that identity verification was about to become a very large, recurring, regulated market, and that the team’s computer-vision edge would matter more inside financial services than in a photo app. It paid off. HyperVerge went on to run KYC checks at scale for financial institutions across India and Southeast Asia, and by the company’s own account has now verified more than a billion identities for a customer base that includes 75+ banks and lenders. The contrast between the two sides of that decision is stark: an eight-figure exit in 2016 that was declined, against a company that a decade later posts ₹149.2 crore of annual revenue and calls itself profitable. The refusal is the reason there is a story to tell at all.

The money behind it

HyperVerge’s funding history is short, which is the whole point. What is notable is how little outside capital sits behind a company of this size.

• Seed round, August 2015: about $1 million from US-based venture investors, reported at the time by Business Standard and YourStory, with New Enterprise Associates (NEA), Naya Ventures and Milliways Ventures among the backers (Tracxn; Business Standard, August 2015).
• Total raised: aggregators including Tracxn record a single seed round and roughly $1 million in disclosed funding; the company itself frames total capital raised at about $1.1 million and describes building “a long-standing bootstrapped company” (Tracxn; company).
• Early angel and mentor: Zoho founder Sridhar Vembu became an early client and investor and an informal mentor, according to Sify’s profile — a notable name given Zoho’s own bootstrapped-to-scale reputation.
• Valuation: no independently verified valuation is on the public record, so this piece does not put a number on the company’s worth. Being bootstrapped, it has not had the priced venture rounds that usually generate one.

Note on the record: some third-party summaries describe HyperVerge as backed by later-stage Indian funds, but across the filings and reports opened for this article the only documented external round is the 2015 seed. Where a claim could not be traced to a source, it has been left out.

How it makes money

HyperVerge earns the way most enterprise software-as-a-service firms do, with a usage twist that suits identity checks.

• Money in: recurring software fees from enterprise contracts, typically tied to volume — the number of verifications, KYC checks or onboarding journeys a client runs. The company describes handling large daily check volumes (an older Sify profile cited around one million KYC checks a day), so per-transaction pricing scales with a client’s own growth.
• The recurring-revenue signal: the company states more than $19 million in annual recurring revenue (ARR), company-stated and echoed by YourStory in 2025. Its Indian-entity operating revenue of ₹149.2 crore in FY25 is the figure that actually appears in statutory filings.
• Where the margin sits: the cost of an extra verification is mostly compute and model-serving, so gross margins on incremental volume are high once a client is live; the heavy costs are engineering talent, sales into regulated buyers, and staying compliant across jurisdictions.
• The part people get wrong: this is not a one-time software licence. KYC and onboarding are continuous, regulation-driven activities, so revenue compounds as clients grow and as rules tighten — which is exactly why a capital-light company could keep scaling without repeatedly returning to investors.

The numbers

The statutory picture below is for HyperVerge Technologies Private Limited, the Indian operating entity (CIN U72400TN2013PTC094387). Figures are in ₹ crore.

• FY25 revenue: ₹149.2 crore per Inc42, corroborated at ₹149.16 crore by TheCompanyCheck — two independent reads of the same filing, up about 16.8–17% on FY24 (Inc42; TheCompanyCheck).
• FY24 revenue: ₹127.7 crore, implied by the FY25 growth rate (Inc42).
• Profit direction: Tofler’s free filing summary shows FY25 net profit down about 87% and EBITDA down about 85% year on year, after an FY24 in which EBITDA had risen about 54.8% and net worth about 44.9% — so the company stayed in the black but profitability compressed sharply in FY25 (Tofler).
• Capital base: paid-up capital of about ₹99.99 lakh against authorised capital of ₹1.5 crore, consistent with a firm that funded growth from operations rather than equity (Tofler).

The exact rupee profit figures for FY24 and FY25 sit behind paid filing-data subscriptions and are not reproduced here; only the audited direction and percentage moves that are visible without a subscription have been used.

Where the money comes from

HyperVerge’s revenue is concentrated by industry and spread by geography — an unusual shape for an Indian SaaS firm.

• By customer type: the heart of the business is regulated financial services — banks, NBFCs, digital lenders, insurers and crypto/wallet firms — with additional demand from gaming, edtech and logistics (YourStory, 2025; company).
• Named customers: the company lists 200+ customers including Reliance Jio, SBI, Bajaj and Aditya Birla Capital, and says it serves 75+ banks and lenders; an older Sify profile also names Vodafone Idea in India and FE Credit in Vietnam.
• By geography: the client base spans 195+ countries per the company, with documented traction beyond India in Vietnam, the Philippines, Malaysia, Indonesia and Nigeria (company; Sify) — several offices sit outside India, including Palo Alto, New York, Singapore, Jakarta and Ho Chi Minh City.
• The surprise: the biggest single figure in the story, the reported $19 million+ ARR, comes on a capital base of roughly $1 million. Very few SaaS businesses reach that revenue-to-capital ratio, and it, not any one product, is what makes HyperVerge worth studying.

The risks

The same choices that make HyperVerge distinctive also create concrete exposures.

• Profit compression, not just slower growth: FY25 revenue grew about 16.8%, a deceleration for a company of its stage, while net profit and EBITDA fell roughly 85–87% year on year per Tofler’s summary. A capital-light firm cannot lean on a cash pile to absorb a squeeze like that; if the margin fall reflects higher costs to win the next tranche of growth, the next year’s filing matters a great deal.
• Regulatory dependence: demand is created by KYC and anti-money-laundering mandates. That is a tailwind when rules tighten, but the Reserve Bank of India and other regulators can also change permitted verification methods (for example around video-KYC or Aadhaar-based checks), and a shift can force costly product rework or shrink a use case overnight.
• Concentration in financial services: the revenue base leans heavily on banks, lenders and fintechs. A downturn in Indian digital lending, or tighter onboarding norms that reduce new-customer volumes, would flow straight into HyperVerge’s usage-based fees, since it is paid per verification.
• Crowded market: identity verification and KYC in India is contested by well-funded rivals such as Signzy, IDfy, Digio and others, several of which have raised far more venture capital. A bootstrapped balance sheet is a discipline, but it can also limit how aggressively HyperVerge can price, market or out-hire better-capitalised competitors.

The takeaway

The transferable lesson from HyperVerge is not “stay bootstrapped.” It is that the scarcest resource for a young company is not money but the willingness to abandon a beautiful idea for a boring, mandatory one. The team’s cleverest work — rail inspection, a photo app, a calorie-reading health app — produced almost no revenue. The business only appeared when they pointed the same technology at a problem regulators forced every bank to solve, and then had the nerve to refuse a life-changing buyout to see it through. Capital efficiency followed from that clarity; it did not cause it. For any founder, the useful question is less “how much can we raise” and more “which unglamorous, unavoidable problem are we uniquely built to solve” — and whether they would turn down the easy exit to find out.

Frequently asked questions

What does HyperVerge do?

HyperVerge sells AI-based identity verification, KYC, KYB and customer-onboarding software to businesses — mainly banks, lenders, insurers and fintechs — so they can confirm a new customer’s identity in seconds and block fraud while staying compliant.

How much money has HyperVerge raised?

About $1 million, in a single seed round in August 2015 backed by investors including NEA, Naya Ventures and Milliways Ventures. The company describes itself as bootstrapped and puts total capital raised at roughly $1.1 million (Tracxn; company).

What is HyperVerge’s revenue?

Its Indian entity reported operating revenue of ₹149.2 crore for FY25 (year to 31 March 2025), up about 16.8% on FY24’s ₹127.7 crore, per Inc42 and corroborated by TheCompanyCheck. The company separately states more than $19 million in annual recurring revenue.

Is HyperVerge profitable?

Yes, but with a caveat. Tofler’s filing summary shows the company stayed profitable in FY25 while net profit fell about 87% and EBITDA about 85% year on year, after a stronger FY24. Exact rupee profit figures are behind paid filing subscriptions.

Who founded HyperVerge and where is it based?

It was founded by five IIT Madras engineers — Kedar Kulkarni (CEO), Vignesh Krishnakumar (CTO), Kishore Natarajan, Praveen Kumar and Saivenkatesh Ashokkumar — and incorporated in December 2013. Its registered office is at IITM Research Park in Chennai, with operational headquarters in Bengaluru and offices abroad.

Sources

Figures are as of September 2026. Currency converted at $1 ≈ ₹96.0 as of 18 September 2026 (Trading Economics).

• Inc42 — HyperVerge company profile (revenue, FY25 ₹149.2 crore, headcount), September 2026
• TheCompanyCheck — Hyperverge Technologies Private Limited (FY25 revenue ₹149.16 crore, up ~17%), September 2026
• Tofler — Hyperverge Technologies Private Limited, CIN U72400TN2013PTC094387 (incorporation, registered office, directors, paid-up capital, FY25 profit/EBITDA direction), September 2026
• Tracxn — HyperVerge company and legal-entity profiles (funding rounds, investors, incorporation), September 2026
• Business Standard — “HyperVerge raises $1 million from venture capital firms,” August 2015
• YourStory — “22 year olds’ deep learning startup HyperVerge just raised $1M,” August 2015; and “Kedar Kulkarni: a journey from campus to conscious AI,” September 2025
• Sify — “HyperVerge: how an IIT Madras start-up built world’s leading image-AI tech” (origin, pivot, December 2016 acquisition offer, early customers, Sridhar Vembu), 2025
• HyperVerge — official About page (ARR, identities verified, countries, customers, founders, offices, NIST rankings), September 2026

Found an error? Write to us and we’ll correct it in the open, dated, on the piece.