Nipun Dave

AI Engineer with 4+ years of experience building AI applications, LLM-powered solutions, and Agentic AI systems that automate complex business workflows. Experienced in designing RAG pipelines, tool-using AI agents, multi-agent systems, and agentic workflows using Python, LangChain, LangGraph, CrewAI, and MCP.

AI Engineer at LTM (2023-10 – Present)

• Built a multi-agent supply-chain exception analysis system using Python and LangGraph, with four parallel Inventory, Procurement, Logistics, and Warehouse agents producing structured, evidence-grounded findings.
• Implemented MCP-based tool integration over SQLite and deterministic validation/correlation workflows to verify agent outputs and connect cross-domain exceptions using predefined business relationships.
• Developed Investigation and Recommendation agents to assess correlated evidence, distinguish correlation from causation, and generate evidence-backed recommendations requiring approval.

AI Engineer at LTM (Present)

• Developed an LLM-powered knowledge assistant for querying enterprise documents such as supply-chain SOPs, contracts, Incoterms, and vendor manuals.
• Built a RAG pipeline using LangChain for document ingestion, recursive chunking, embedding generation, vector storage, and semantic retrieval.
• Implemented retrieval-grounded prompting to generate answers based on relevant document context and reduce unsupported responses.
• Added source metadata to retrieved chunks to make generated answers traceable to the underlying documents.

AI Engineer at LTM (Present)

• Developed a computer-vision solution to analyze product presence and shelf-level product distribution from retail shelf images.
• Built an image-processing pipeline for preprocessing shelf images and detecting product instances using computer-vision techniques.
• Used detected product instances and category-level counts to derive product-presence and shelf-share indicators.

ETL Developer at LTM (2022-04 – 2023-10)

• Developed regex-based ETL pipelines to clean, normalize, and generate standardized master data from raw PlantGraph data.
• Implemented pattern-similarity matching and tag transformation to map standardized master data with SAP, CREDO, and PHD source systems, automating engineering data reconciliation.

M. Tech in Data Science & Eng. – BITS Pilani (WILP) (2022-04)

B. Tech in Computer Science & Eng. – VIT University (2016-05)