AI Engineer passionate about building practical LLM-powered applications and exploring how AI systems can be made more efficient. Currently, I'm working on training and fine-tuning models to reduce LLM token costs while maintaining performance. My experience includes building end-to-end RAG pipelines, natural language-to-SQL chatbots, semantic search systems, and multi-stage LLM workflows. I work primarily with Python, LangChain, vector databases, SQL/NoSQL databases, and modern backend technologies. Beyond AI, I enjoy writing about LLMs, RAG, agentic AI, and system design, sharing lessons learned from real-world projects and experiments. I also have research work published at WWW RelWeb 2025 and am always looking to learn and build something new.