Showcasing AI/ML work and open-source contributions
A comprehensive framework for fine-tuning large language models with support for multiple architectures, efficient training techniques (LoRA, QLoRA), and deployment utilities. Built with PyTorch and HuggingFace Transformers.
Retrieval-Augmented Generation system for answering questions from large document collections. Implements vector search with FAISS, semantic chunking, and context-aware answer generation using LLMs.
End-to-end computer vision pipeline using state-of-the-art object detection models (YOLO, Faster R-CNN). Includes data preprocessing, model training, evaluation, and deployment with REST API.
Production-ready MLOps template with CI/CD pipelines, model versioning, experiment tracking, and monitoring. Supports multiple cloud platforms and includes best practices for ML deployment.
Fast and accurate sentiment analysis API using fine-tuned transformer models. Supports multiple languages, batch processing, and real-time inference with sub-100ms latency.
I'm always interested in working on innovative AI projects. If you have an idea or want to discuss potential collaborations, feel free to reach out through GitHub or LinkedIn.
Contact Me →