Prasad K.

I’m an AI Applied AI Researcher with a strong focus on building real-world GenAI and machine learning systems that actually work in real life use cases. At Fujitsu Research, I’ve been working on some pretty exciting stuff like building intelligent agents using Hybrid RAG (dense + keyword) with Neo4j knowledge graph for enterprise chatbots, creating GenAI platforms that enhance passenger experience at Heathrow Airport, and even designing vision-language models for document understanding and zero-shot HOI detection. I love working across the stack, whether it’s building NLP pipelines, designing smart retrieval strategies, fine-tuning LLMs, or deploying models with Docker, MLflow, and AWS. Some of the things I’ve worked on recently: - LLM-based chatbot agents that reason over knowledge graphs - A semantic search platform using both static + dynamic data - Multimodal document QA benchmarks and lightweight visual modules - State-of-the-art zero-shot detection models with CLIP I did my M.Tech from IIT Bombay (9.49 CGPA) and have been blending research with real-world applications ever since. Always open to new ideas, collaborations, and opportunities, especially around GenAI, LLMs, and applied ML. Let’s connect!