Results-driven AI Engineer specializing in end-to-end, CPU-optimized Computer Vision and Generative AI systems, with
pro

github

Shubham 333k  • Peerlist

DefectX AI is a production-ready, explainable AI system for automated manufacturing quality control. It combines state-of-the-art YOLO object detection with Gradient-weighted Class Activation Mapping (Grad-CAM) to not only identify defects in manufacturing products but also provide visual explanations for its decisions, building trust in AI-powered inspection systems. 🎯 Key Differentiator: Explainable AI (XAI) Unlike traditional defect detection systems that operate as "black boxes," DefectX AI provides Grad-CAM heatmaps that highlight the exact regions of an image that influenced the model's decision. This transparency is critical for: Quality Assurance Teams: Understanding why a product was flagged Regulatory Compliance: Demonstrating AI decision-making processes Continuous Improvement: Identifying patterns in false positives

DefectX AI - Explainable AI Quality Control S

Enterprise RAG Chatbot (internally configured as the Enterprise Knowledge Assistant) is a self-hostable Retrieval-Augmented Generation platform built for organizations that need employees to query internal documents — policies, engineering docs, HR handbooks, sales collateral — without leaking data across departments or paying for a dedicated on-prem GPU. It solves three problems that generic "chat with your PDF" demos usually ignore: ✅ Department-level data isolation — an HR employee's query never surfaces Engineering-only documents, because access filters are applied at the vector database query stage, before the LLM ever sees the retrieved chunks. ✅ No vendor lock-in on the LLM — the generation layer is provider-agnostic (OpenAI / Gemini / Groq), so you can run entirely on free-tier APIs (Gemini, Groq) or swap to OpenAI for production quality without touching the retrieval pipeline. ✅ Runs on modest hardware — embeddings are generated locally with a lightweight sentence-transformer model, so the only network calls are to the LLM API; no GPU, no 16GB+ RAM requirement.

Enterprise RAG Chatbot

MedNavigator-AI is a production-grade, AI-powered assistant built specifically for the healthcare domain. It solves the core problem clinicians face daily — the overwhelming volume of medical literature and the difficulty of extracting relevant, evidence-based answers quickly and safely. By combining Google Gemini 2.5 as the language engine with a Retrieval-Augmented Generation (RAG) pipeline, MedNavigator-AI retrieves the most relevant chunks from your private medical knowledge base before generating a response. This approach: ✅ Reduces hallucinations by grounding answers in real source documents ✅ Provides citations for every answer so clinicians can verify the source ✅ Keeps data private — your knowledge base stays on your infrastructure ✅ Maintains HIPAA compliance through PHI de-identification and audit logging

MedNavigator-AI

UrbanFlow is a cutting-edge, AI-driven computer vision system designed for intelligent traffic monitoring and adaptive signal control. Developed as a BTech Major Project, it offers a comprehensive, standalone pipeline for real-time analysis of traffic flow, detection of violations, and dynamic adjustment of traffic signals. Optimized for CPU usage, UrbanFlow makes advanced AI traffic management accessible for academic research and practical deployment in smart city initiatives.

Shubham 333k

Results-driven AI Engineer specializing in end-to-end, CPU-optimized Computer Vision and Generative AI systems, with pro