Agentic Offensive Security at Scale
We're @Huzaifa Ahmad, @Ahmad Khan, and @Prama Yudhistira (Parameter W26). We built AI agents that automate penetration testing.
In the past few weeks, our agents found critical vulnerabilities in dozens of YC companies (ones you’ve heard of): SQL injection exposing billions of records, demonstrated a PoC worm that could infect entire networks, gained access to hundreds of codebases, and more. Through this, we have prevented an estimated $3B+ in potential damages (based on exposed record counts and IBM's Cost of a Data Breach benchmarks).
The problem: Companies lose $10T+ annually to preventable vulnerabilities, yet most pentest once a year because it's manual, expensive, and slow. By the time you get the PDF, your codebase has changed.
And that gap is widening. In September 2025, Anthropic caught Chinese state-sponsored hackers using AI to run 80-90% of an attack autonomously — thousands of requests per second, speeds no human team could match. The attackers have AI now, and defenders need it too.
What we built: AI agents that run continuous penetration tests against your apps and infrastructure. They find vulnerabilities, chain exploits together, and deliver actionable findings with reproduction steps — at a fraction of the cost of traditional pentesting.
Want to see what we find in your systems? Book a discovery call: hex.co
Visit us at hex.co | founders@hex.co
