Backend Developer | AI/ML Engineer Final year Electronics & Telecommunication student at SITS, Narhe, Pune. I build production-grade backend systems and AI infrastructure.
-
ARGUS — Multi-model AI orchestration framework routing queries across Gemini, OpenAI, and Claude in parallel with async dispatch and a composite eval harness scoring quality, cost, and latency.
-
DocuMind — Production RAG platform with two-stage FAISS and cross-encoder retrieval, BYOK LLM routing, DeepEval-based generation scoring, and a 31-endpoint Django/FastAPI backend.
-
Meridian — Full-stack productivity platform on React Native, FastAPI, PostgreSQL, Redis, and Supabase Realtime with two-way Google Calendar and Outlook sync and Whisper API transcription.
-
DMP (in progress) — Distributed LLM training daemon. BitTorrent-style, DHT peer discovery, coordinate-wise median gradient aggregation, node trust scoring, no central server.
Backend: Python, Django, FastAPI, DRF, PostgreSQL, Redis, Celery, WebSockets, JWT Auth, Supabase
AI/ML: RAG, LLMs (OpenAI, Gemini, Llama, Claude), Vector Embeddings, pgvector, FAISS, DeepEval, Whisper API, AI Agents
Frontend/Mobile: React Native, Expo, React, REST API Integration
Tools: Docker, Git, Linux, Postman, Power BI
Languages: Python, C, C++, Java, SQL, HTML/CSS
- Data Science & Analytics Intern — Zidio Development (May 2024 – Aug 2024)
- Power BI Intern — KasNet Technologies (Jan 2025 – Mar 2025)
- Email: joshi.narayan2004@gmail.com
- LinkedIn: narayan-joshi-85227425a
- GitHub: Narayan201120