Pranav Shende – Full Stack Developer & Forward Deployed Engineer
Based in Nagpur, India • Building scalable backend systems, AI-powered applications, RAG solutions, and production software.
Professional Profile
Pranav Shende is a Full Stack Developer and Forward Deployed Engineer based in Nagpur, India. He specializes in backend engineering, AI-powered applications, enterprise software, Retrieval-Augmented Generation (RAG), and scalable system architecture.
His engineering work includes enterprise intelligence platforms (Jarvis AI), clinical AI systems (Electropathy AI), distributed food delivery logistics (Tastifyy), civic governance applications (JanSampark), clean-tech big data analytics (Solar Intelligence), and edge veterinary diagnostics (PashuRakshak).
His technical ecosystem includes React 19, Node.js, PostgreSQL, Python, Prisma ORM, Fastify, Supabase, Apache Kafka, Redis, Model Context Protocol (MCP v1.0.4), pgvector, and Google Gemini 2.5 Pro/Flash.
Autonomous enterprise intelligence platform unifying fragmented operations across ERP, warehouse management (WMS), and finance systems into a centralized intelligence layer. Features official Model Context Protocol (MCP v1.0.4) streamable HTTP and stdio transports, zero-hallucination deterministic SQL calculations across a 35-table PostgreSQL star schema, hybrid pgvector RAG with 749+ chunks, and dynamic Google Gemini 2.5 Pro routing.
Tech: Google Gemini 2.5 Pro • Model Context Protocol (MCP) • Fastify Gateway • pgvector • Supabase • BullMQ • TypeScript
Agentic clinical decision-support and medical knowledge engine built for Dr. Ashish Wandhare with pgvector semantic search and multilingual RAG copilot in English, Hindi & Marathi. Strict zero-hallucination guardrails grounded in verified literature.
Multi-role food delivery platform featuring Apache Kafka event-driven order pipelines, Redis geospatial caching for instant driver proximity dispatch, React 19 web portal, Expo mobile app, and live Socket.io tracking.
Offline-first Edge AI veterinary diagnostics system with quantized convolutional neural networks (TensorFlow Lite INT8) for sub-120ms livestock pathology detection in zero-connectivity rural farms.
CleanTech big data intelligence platform analyzing 10+ years (2014–2024) of solar energy across 36+ states/UTs covering 81.8+ GW with automated ETL pipelines, GIS heatmaps, and polynomial regression forecasting to 2030.