Building CliniChat - AI-powered healthcare management for clinics and dental practices. We're rethinking how healthcare providers handle patient communication, scheduling, and day-to-day operations.
Previously built Dextego as CTO & Co-founder - real-time AI coaching for enterprise sales teams.
Before startups, I worked on conversational AI at Skit.ai, Frrole AI, and Humantic AI.
I spend most of my time at the intersection of AI systems, real-time infrastructure, and healthcare technology.
The problems I find most interesting involve building software that needs to be simultaneously intelligent, fast, and trustworthy - especially in domains where reliability isn't optional.
On AI in production - The model is rarely the bottleneck. What matters is everything around it: latency, reliability, cost, evaluation. Production AI is fundamentally a systems problem.
On healthcare software - Healthcare moves slowly for good reasons. The challenge is building modern, intuitive experiences without compromising the trust and safety that patients and providers depend on.
On real-time systems - Real-time isn't about speed; it's about predictability. Tail latency matters more than throughput, and architectural choices made early determine what's possible later.
On building companies - The best technical decisions are the ones that buy you time to learn. Optimize for iteration speed early, correctness later - but know when to flip that.