Shaikh Hasibul Islam

I’m a performance-driven AI Engineer with a strong Computer Science foundation and published research in Deep Learning. I specialize in building scalable Generative AI applications, with a focus on RAG architectures and LLM workflow optimization.

In my projects, I engineered a PDF Chatbot using Retrieval-Augmented Generation, enabling semantic search across documents with high contextual accuracy. I also optimized backend memory footprint by migrating from local sentence-transformers to the Hugging Face Inference API, improving deployability in resource-constrained environments.

I further built an autonomous AI Agent with LangGraph and LangChain, including multi-step reasoning and real-time web research through Tavily. I deploy these systems with a FastAPI backend and Streamlit frontends, and I’m motivated by turning AI ideas into production-ready, maintainable experiences.