I buildmulti-agent AI,full-stack platforms& data pipelines.
Kumar Priyam
B.Tech Computer Science at NIT Delhi, Class of 2027. I work end to end — from RAG architectures and reinforcement learning to production ETL.
Selected Work
(4)- 01
JobMatch: Semantic Job-Matching Platform
Multi-tenant platform ranking remote jobs against your résumé via Gemini vector embeddings and pgvector.
Next.js 15FastAPISupabase/pgvector - 02
ClinicQ: WhatsApp Virtual Queue for Clinics
WhatsApp-native token queue for single-doctor Indian clinics: patients book, track live ETA, and check in with no app, while reception runs the queue from a one-thumb PWA.
WhatsApp Cloud APIFastAPINext.js 14 PWA - 03
MARL-MAPS: Dynamic Multi-Agent RL for Optimized RAG
Decentralized Reinforcement Learning policy architecture eliminating RAG Context Tax.
PythonMulti-Agent RLRAG-DDR - 04
Intelligent Interview System
Low-latency candidate evaluation platform utilizing multi-agent AI consensus pipelines.
MERN StackGoogle Gemini APIWebRTC
— What I do
AI Engineering
Multi-agent RL, RAG architectures, LoRA fine-tuning and orchestration policies — MARL-MAPS cut redundant retrieval by 91%.
Software Engineering
Full-stack platforms on FastAPI, Next.js and the MERN stack — real-time systems with sub-70ms sync and 46% lower API latency.
Data Engineering
Multi-tenant ETL on Apache Airflow with strict isolation, schema validation and automated fault classification at 100K+ records/day.