Sumit is a key contributor within BlackRock's Aladdin Data, responsible for engineering the scalable infrastructure that underpins global financial intelligence. His work focuses on optimizing high-throughput data pipelines involving AI based processing and establishing the governance framework essential for securing and managing vast, critical datasets within a highly regulated environment. He also presented his ideas on SLM-driven agentic AI at PyCon US 2025 in Pittsburgh.
He leverages a deep background in Generative AI and Advanced Data Science to future-proof the platform. Previously, as an AI Engineer, Sumit specialized in Large Language Models (LLMs), designing and deploying complex, agent-based workflows using LangChain and LangGraph. He has hands-on experience with advanced fine-tuning techniques, including RLHF, QLoRA, and PEFT, alongside building robust retrieval systems using RAG.
Sumit's technical acumen extends to performance evaluation using metrics like RAGAS, GLUE, and MMLU, and integrating specialized tools like GPT-4 Vision and AWS Textract for unstructured data processing. Holding a B.Tech in IT and a Diploma in Data Science from IIT Madras, he combines research-driven insights with practical engineering to build robust, user-centric GenAI solutions that scale.