Sara Manzoor

I'm a full-stack developer and ML engineer exploring the intersection of applied AI and real-world software. My work spans classical ML (XGBoost, LightGBM, Scikit-learn) — including a Kaggle-ranked regression model and a healthcare cost-prediction system built for a national ML competition — through to computer vision (YOLOv8, OpenCV) and deep learning (TensorFlow/Keras) for tasks like sign language recognition. Lately I've been diving into generative AI, RAG pipelines, and LLM-powered applications, including a bilingual AI assistant built with Llama 3 and LangChain that placed Top 10 at a Generative AI hackathon. I'm here to learn from and contribute to the open-source ML community as I go deeper into building production-ready AI systems.