About Me

I’m a Data Scientist and Applied AI Engineer with 4+ years of experience building intelligent systems across Retrieval-Augmented Generation (RAG), NLP, and real-time computer vision. I specialize in designing end-to-end machine learning pipelines — from structured data ingestion and retrieval optimization to lightweight deployment in production environments.

Recently, my work has focused on retrieval system optimization and evaluation. I have built hybrid semantic search pipelines using SBERT, BM25, and vector databases, and validated them using RAGAS metrics to ensure faithfulness, context recall, and grounding quality. I enjoy solving practical AI problems where performance, evaluation, and system design matter as much as model accuracy.

In computer vision, I have engineered CPU-optimized real-time detection and tracking systems using YOLOv8 and ByteTrack, balancing accuracy and performance under deployment constraints. My academic background in Electronics & Communication Engineering and Signal Processing gives me a strong systems-thinking mindset when designing scalable AI solutions.

I focus on building clean, evaluation-driven, production-aware AI systems — not just models, but reliable solutions.