Chetan Saini

I’m a Software Engineer with end-to-end ownership across backend services, AI/LLM features, and cloud infrastructure. I ship production systems that turn LLM capabilities into reliable user experiences.

At Vani.coach, I designed and shipped LLM-powered feedback generation across three rubrics, currently producing ~5,000 feedback generations/month in production. I also built two production real-time voice agents on Pipecat/Daily with Azure Speech (STT/TTS)—including an onboarding agent that replaced human coach-led discovery calls for ~90% of incoming learners. I reduced Azure infrastructure spend 73% ($3,000 → $800/month) by consolidating onto a single VM, right-sizing the database, and removing high-cost feature paths.

Previously at MSG91, I led full stack development of a KnowledgeBase help center, optimizing page load speeds by over 10x and maintaining a 95+ SEO score. I implemented semantic search, built an LLM-based support chatbot, and added tracking for views and reader feedback—always focused on performance, relevance, and measurable outcomes.