Using AI to find security holes in software — including in other AI products — and proving each one is real.
I built an AI-driven penetration testing harness that reasons about the specific application, reproduces every finding before reporting it, and has a dedicated module for AI features. It only ever runs with written authorisation.
Product engineering teams
An authorised penetration test run through the harness — the model handling recon and targeted test generation, with every finding reproduced before it reached the client. Delivered ranked, with fixes attached and a re-test to confirm them.
Contact
I'm open to AI engineering, architecture and training work. Tell me what you're building and what the constraints are — that's usually enough to start.