Hiren Prajapati

I've built AI-powered applications including EDACracker, a Flask-based GenAI analytics platform that automates dataset profiling, insight generation, and machine learning recommendations using LangChain and OpenRouter-hosted LLMs. For FoodLens, I developed an end-to-end food intelligence pipeline using YOLOv8 and a fine-tuned Vision Transformer, achieving 90.77% mAP@50 detection and 92.05% classification accuracy across 21 Indian food categories. I also built its FastAPI and PostgreSQL backend for serving-size-aware nutrition output. I've also created SentinelLine, a production monitoring and anomaly detection system using Isolation Forest and LSTM Autoencoders, with Streamlit and Plotly dashboards for operational insights. I'm an Artificial Intelligence and Machine Learning Engineering student with hands-on experience in Python, FastAPI, Flask, Java, data analysis, computer vision, NLP, and generative AI. I enjoy solving practical problems through scalable software and real-world AI projects.