Built by engineers who watched AI adoption outrun the controls meant to govern it.

We both came up leading engineering teams in regulated industries — healthcare and financial services — where the cost of a data leak isn't a bad headline, it's a HIPAA fine, a breach disclosure, or a patient harmed.

When AI coding assistants and chat tools exploded onto our teams, the conversation always went the same way: developers loved them, and security teams had no idea what was actually being sent. Patient records. Internal financial models. Credentials committed to prompts. The tooling to answer “what just left our network?” simply didn't exist.

We tried to buy the solution. We evaluated every DLP product, every network proxy, every AI-specific monitoring tool on the market. Nothing solved the problem cleanly — they either required SDK rewrites that developers would never accept, produced so much noise that security teams tuned them out, or had no concept of what an AI request actually looked like.

So we built BastionGate: a transparent gateway that sits between your team and every AI provider, inspects every request in real time, and enforces the policies your security team actually needs — without changing how developers work.

The need is immediate. If your team uses Cursor, VS Code, Claude Code, ChatGPT, or any AI tool, data is moving. The question is whether you know what it is.

Joey & David