About
I'm a software engineer with 4+ years of experience building production-grade AI systems and distributed infrastructure. IIT Jodhpur B.Tech 2022. Currently at Avataar.ai working on voice-first AI for rural India — real-time multilingual crop diagnosis and insurance automation. I care deeply about systems that scale and products that matter.
Outside work, I'm building Uplyft — a real-time field force tracking platform available on the App Store and Play Store.
Experience
- ▸Architected a production-grade, voice-first AI farming assistant (Kisan-Vegh) using React/TypeScript, Django + FastAPI backend, and PostgreSQL — real-time audio transcription, multimodal image analysis, and multilingual LLM responses.
- ▸Built a multi-model orchestration pipeline (Sarvam ASR, Dhenu VLM, LLM) with centralized error handling, per-model fallbacks, and graceful degradation on partial failures.
- ▸Built a real-time multi-agent observability dashboard (Socket.io + custom React hooks) with event-driven state transitions and sub-second UI updates.
- ▸Automated PMFBY crop insurance filing triggered by AI-detected anomalies — vision model output parsed and routed through an LLM agent, eliminating manual intervention detection-to-filing.
- ▸Presented at Impact AI Summit 2026 demonstrating real-time multilingual crop diagnosis and insurance automation for rural farmers.
Projects
Uplyft
Real-time field force tracking and management platform — built and shipped solo. Helps businesses monitor, assign, and coordinate field teams with live location tracking and analytics.
Kisan-Vegh / UAI Dashboard
Voice-first AI farming assistant with multilingual crop diagnosis, PMFBY insurance automation, and a real-time multi-agent observability dashboard.
IPAAS — Integration Platform
In-house CRM integration platform supporting 23+ integrations. Replaced Blendr, saving $60K/yr and reducing event setup from days to hours.