Hey, it’s Gio.

If you’ve been following us, you’ve probably seen a few posts announcing our seed round and breaking down our approach to shift-left security in crypto.

Moving forward, I’ll be writing these updates personally.

I want to use this space to speak directly to the engineers, security researchers, security engineers, auditors, and builders who care about shipping secure code — and doing it faster, cheaper, and more reliably than manual audits alone can offer.

At Octane, we build machine learning models that find bugs in codebases before they go live.

I studied CS with a concentration in AI and machine learning at Duke, but dropped out after seeing one too many teams get wrecked by exploits, even after they’d spent six figures on audits.

I learned that lesson the hard way after experiencing some exploits first-hand.

I’ve been obsessed with security ever since.

But the reality is:

Security in crypto still lags behind the speed of development.

Audits help (when they’re done well) but they’re slow, manual, and time-consuming.

Manual security cycles aren’t just slow, they’re unpredictable. Lead times can stretch to months for a manual audit. Quality varies depending on the individual auditor. And six-figure audit fees don’t always catch all of the high-impact issues. I’ve seen teams burned by missed vulnerabilities even after paying top dollar.

At Octane, we’re trying to fix that.

What We’re Building at Octane

Octane is an always-on AI security platform for crypto teams.

Instead of relying purely on point-in-time audits, we help teams shift security left by catching vulnerabilities throughout the development lifecycle, from internal reviews to post-audit checks.

This shift-left approach means embedding security from day one instead of tacking it on right before launch.

Our models are trained on real-world exploits, code reviews and open-source code. They can surface issues like reentrancy exploits, missing slippage safeguards, signature manipulation, access-control vulnerabilities, etc.

Teams like Circle, Decent, and Redstone Oracles already use Octane to catch criticals early.

And we just raised a $6.75M seed round from Archetype, Winklevoss Capital, Circle, and others to keep building.

Why This Newsletter Exists

I want to use this space to:

• Share deep dives on real hacks (and how they could’ve been prevented)

• Talk about what’s working in AI-based security and where things still need work

• Push forward the conversation on how security actually scales in crypto

This isn’t content for beginners.

It’s for engineers, auditors, and founders who are in the weeds and care about getting this stuff right.

In future issues, I’ll also highlight real vulnerabilities Octane has caught and explain what made them dangerous in plain English.

If you’re curious to see what AI can do in this space, or you’re building in crypto and want a second set of eyes on your code, I’d love to hear from you.

More soon,
Gio

Ready to Secure Your Smart Contracts?

Deploy with confidence by adding Octane’s AI security to your CI/CD pipeline. Schedule a live demo to see how we deliver 24/7 offensive intelligence and real-time vulnerability detection.