K boopathi

I’m a Full Stack Engineer focused on production-grade generative AI and real-time voice systems. I build end-to-end experiences that turn LLM capabilities into reliable, multi-step workflows and useful customer-facing voice agents.

In my current role, I designed a workflow builder enabling LLMs to run complex multi-step voice workflows, and I built an automated code review agent to reduce manual effort and improve quality. I also developed a real-time AI voice agent using Pipecat and open-source models, and delivered a proof of concept Smart Call Agent with Twilio and ElevenLabs.

Previously, I helped migrate production from Heroku to Kubernetes with zero-downtime deployments, health probes, and incident-focused debugging. I improved performance across the stack—cutting frontend build times by 70% (Webpack → Rspack), accelerating TypeScript compilation using Rust-based SWC, and mitigating Redis overload during a DDoS by reducing Redis load by 90%.

I enjoy scaling systems and making problems feel solvable, whether that’s designing scalable REST APIs, integrating AI models into real workflows, or writing about debugging and infrastructure challenges on Dev.to and Medium. I’ve also earned recognition through hackathon wins (like DinoDial) and internal culture awards, and I’m proud of delivering practical results with strong engineering discipline.