Devansh Khandekar

I’m an Applied ML / AI Engineer and Data Scientist with 6+ years of practical experience building production-ready AI systems across Computer Vision, NLP, RAG, AI agents, time-series modelling, and MLOps-style deployment.

I specialize in object detection, image classification, segmentation, OCR, face recognition, tracking, defect inspection, real-time video analytics, semantic search, chatbots, and AI-agent workflows. I have built YOLO/OpenCV-based vision systems, PaddleOCR/ANPR pipelines, RAG applications with pgvector/Postgres, LangGraph/LangChain agents, ASR/voice-search services, and production APIs using FastAPI/Flask, Docker, and cloud infrastructure.

My focus is not just model training, but complete deployable solutions: data pipelines, model tuning, inference APIs, dashboards, logging, documentation, and client-ready handoff.