Building production LLM systems β eval frameworks, voice agents, healthcare automation.
Previously: 98% accuracy LLM evals across 120K medical procedures at a YC W23 startup.
- Real-time voice + LLM systems at Mynaksh β WebRTC/VoIP infrastructure, agentic conversation flows, AI-driven onboarding.
- Previously at Mantys (YC W23) β production LLM eval framework reaching 98% extraction accuracy, prior-auth automation across 120K medical procedures, $2.3M in claims recovered.
Currently going deep on: agent observability, voice infra, healthcare AI.
Open to: founding/senior engineer roles at AI-native companies. Especially healthcare AI, voice agents, and LLM evals/observability.