Horizon3
NightDragon
NEA
Acrew Capital
Blue Cloud VenturesDDemeter Group
EDBI
PSG
SAIC
Sapphire Ventures
Craft Ventures
Prosperity7 Ventures
Qualcomm Ventures
Ridge Ventures
SignalFireThe Story
Horizon3 has raised $250 million, about ₹2,400 crore, in a Series E round at a valuation the company puts at more than $2 billion. The round was announced on 3 August 2026 and was co-led by existing investors NightDragon and NEA. Seven new investors joined: Acrew, Blue Cloud Ventures, Demeter Group, EDBI, PSG, SAIC and Sapphire. Five returning backers participated, including Craft Ventures, Prosperity7 Ventures, Qualcomm Ventures, Ridge Ventures and SignalFire. Dave DeWalt, founder of NightDragon and a former chief executive of FireEye and McAfee, joins the board alongside NightDragon managing director Morgan Kyauk. The company closed its Series D in June 2025 at a $650 million valuation, making this roughly a tripling in fourteen months. Total funding to date stands at $428.5 million. The capital is going towards international expansion, with offices opened in Amsterdam in June 2026 and further sites planned in Australia and Singapore, alongside building out a partner network, sales operations and research spending. The San Francisco company was founded in 2019 by Snehal Antani and Anthony Pillitiere, who met while serving at Joint Special Operations Command. Antani was previously chief technology officer at Splunk. Its NodeZero platform performs autonomous penetration testing against live production systems. Several figures attached to the announcement come from the company rather than an independent source. Horizon3 says it approached $100 million in annual recurring revenue last year with 120% year-on-year growth, that it has run 310,000 production security tests without disruption, and that it has spent roughly $100 million on research and development. None of these have been independently verified. Customer counts vary across accounts of the round, ranging from more than 6,500 organisations to between 7,200 and 7,300. The company's own statement says more than 7,000.
Why It Matters
Penetration testing has historically been a consulting engagement. A firm is hired, people spend a few weeks probing a slice of the network, a report arrives, and the exercise repeats a year later. The economics of that model cap it: the work is priced by the hour and the coverage is limited by how much a team can examine before the budget runs out. Companies end up testing a small fraction of their infrastructure once a year. Horizon3 sells the same outcome as software. NodeZero runs against live systems without taking them offline, chains together weak credentials and misconfigurations into a working attack path, and re-tests after the fix to confirm the path is closed. Because it is software, running it monthly costs roughly what running it annually costs. That is the whole commercial argument, and it is why the company describes its competition as the incumbent model rather than other vendors. The cost structure is heavy on research rather than delivery. Horizon3 says it has spent around $100 million building systems intended to behave predictably inside production environments, which is the hard part: an autonomous attacker that breaks a hospital network is worse than no test at all. What the announcement does not settle is whether the economics work. Recurring revenue approaching $100 million with 120% growth is a company statement, unaudited, and stated as of last year rather than today. Nothing has been disclosed about gross margin, net revenue retention, customer concentration or whether the business is profitable. A valuation above $2 billion against that figure is roughly 20 times revenue, which prices continued growth at the current rate rather than the growth already delivered.
The Strategic Read
The market assumption being underwritten is that security validation stops being an audit and becomes a running process, and that the thing doing the running is software rather than people. If that holds, the budget line moves from professional services, which is priced per engagement and grows with headcount, to software, which is priced per seat or per asset and does not. That reframing is what justifies the multiple. At roughly 20 times a claimed $100 million in recurring revenue, the price assumes Horizon3 captures a meaningful share of a spend category that mostly does not exist yet as software. The company's own framing supports this reading: it says the competition is the existing annual-test model rather than rival vendors. The moat claim is data. Horizon3 argues that 310,000 tests executed in production give it training material a general-purpose model cannot obtain, because nobody lets an untested system attack a live hospital network. That argument is stronger than a typical data-moat claim, since the data is generated by the product in environments competitors cannot access. It is weaker than it sounds in one respect: the value of proprietary attack data depends on how much better it makes the system than a frontier model with public exploit knowledge, and that gap is not something either side has published. The composition of the syndicate says where the demand is expected to come from. A Singapore state investor, a US defence contractor and a chipmaker's venture arm are not financial investors making a software bet. They are buying proximity to sovereign and critical-infrastructure security spending. The risk sits in the product's central promise. Everything rests on autonomous software attacking production systems and breaking nothing, and the claim is a perfect record across 310,000 tests. That record is an asset until the first time it is not. A single disruptive incident at a bank or a hospital would not merely be a bad quarter; it would undercut the specific proposition the company has spent six years and $428.5 million establishing.
For daily, sharp analysis of the biggest moves in the Indian business and startup ecosystem, follow StartupFox.
