I built a tool to make EU AI Act compliance a little less painful for developers.
Hey everyone š
Iām Mohid, founder of Diliga.
While building AI products myself, I started looking into what the EU AI Act actually means for developers and small teams.
What surprised me was how quickly compliance turns into a documentation problem.
You might know what your AI system does, but then you need to document things like risk management, technical details, transparency requirements, human oversight, and other information depending on the system.
And for a small team, figuring all of that out from scratch isn't exactly fun.
So I built Diliga.
The basic idea is simple:
ā Answer questions about your AI system
ā Diliga analyzes the information
ā It generates a structured compliance dossier
ā You can review and edit everything
ā Keep the documentation updated as your system changes
I built the whole thing myself, from the product and AI pipeline to the document generation side.
I'm still working on it, so I'm especially interested in feedback from developers.
A few things I'm trying to figure out:
How much of the compliance workflow should actually be automated?
What information would developers want to see explained rather than just generated?
Would you trust an AI-generated first draft for compliance documentation?
What would make a tool like this genuinely useful in a real development workflow?
Diliga isn't legal advice and isn't intended to replace a lawyer. The goal is to make the tedious documentation part easier and give teams a solid starting point.
If you're building AI systems and have looked into the EU AI Act, I'd love to hear how you're currently handling compliance.
Top comments (0)