Dhruv Shah

Meet Dhruv

1 following

I test AI agents and LLM apps for the failures demos hide: prompt injection, unsafe tool calls, hallucinations. You get a clear report on what broke, how to reproduce it, and what to fix first. I also build Power BI/Python dashboards. Built Sentinel, an MCP security benchmark covering 6 attack classes. B.Tech EE, Nirma University. Send me your AI product and I'll name 3 issues to test first, free.