A lightweight observability tool for visualizing and comparing RAG retrieval strategies. Features real-time embedding visualization and side-by-side performance metrics.

Edge-native MLOps pipeline for computer vision tasks, featuring automated deployment and resource-efficient inference

A data preprocessing pipeline for TCAS admission data. This project leverages an LLM (Typhoon AI) for advanced, fine-grained text filtering and comparison against traditional Regex methods. Feature…

End-to-end event-driven architecture for real-time semantic video search. Orchestrates YOLOv8 and CLIP models via Kafka for vector embedding generation, stored in Qdrant, and provisioned on AWS usi…

MLOps producer hub for medicine pack identification. Built with Streamlit, YOLOv8-segmentation, and DINOv2. Featuring automated vector database synchronization with AWS S3 and activity logging system

End-to-end MLOps pipeline for real-time pill identification on Edge devices using Deep Metric Learning, Knowledge Distillation, and Vector Search.

Customer Churn End2End Machine learning project