I'm the builder of this and its taken a few iterations to get to where it's at today. Current landscape of regulatory compliance work is so manual and time consuming for critical infrastructure industries, that was the glaring problem that birthed this idea.

I think the chat interface we all use day to day isn't going away and quite possibly the next interface of how we interact with software and agentic layers of getting stuff done.

I managed to aggregate a growing set of regulatory bodies and source material into one db, embed and chunk it, use a not so novel architecture to reason and generate a response with those sources. Using this to then also complete tasks that a real compliance manager would in the office like permit reviews, gap analysis, audits and more.

Real time updating to make sure the knowledge base is as current as possible, getting feedback for access to everything via API for existing software stacks so thats cool ig, but honestly excited to keep building this and see where it goes.

I’d especially appreciate feedback on: - whether the chain-of-thought on responses are useful - whether the answers are actually grounded enough - what sources are missing - what parts of the experience feel too slow or confusing

Happy to answer any q's at nico [@] tryrival dot ai