How each project got built
Problem, approach, and results for the AI systems Sameer Singh has shipped — from an agentic commute planner to a grounded RAG pipeline.
-
01
Saarthi AI
Agentic AI · Commute Risk PredictionHow a 4-person team built a proactive commute-planning agent for Lucknow that uses MongoDB MCP as agent memory, for the Google Cloud Rapid Agent Hackathon (MongoDB Partner Track).
-
02
Scholar_Mate AI
RAG System · Academic Notes AssistantAn end-to-end RAG system for document-grounded Q&A over typed, scanned, and handwritten PDF notes — with OCR fallback and prompting strict enough to say "I don't know" instead of hallucinating.
-
03
RecruitRank AI
Ranking Engine · Redrob Data & AI ChallengeA hybrid candidate-ranking engine built for the Redrob Data & AI Challenge, producing an auditable top-100 shortlist without relying on keyword matching.
-
04
Chest X-Ray Classifier
Computer Vision · Transfer LearningFine-tuning ResNet50 to classify pneumonia from chest X-rays — taking accuracy from 66.7% to 90.1% through transfer learning, differential learning rates, and data augmentation, with Grad-CAM for interpretability.