BeingHuman

AI Engineer with 3+ years of experience building production-grade Generative AI, Agentic AI, RAG, and Machine Learning solutions. Experienced in developing scalable AI systems using LangGraph, FastAPI, AWS, and LLMs. Skilled in multi-agent orchestration, intelligent retrieval, and enterprise AI applications, collaborating with cross-functional teams to deliver solutions that drive automation and business impact.

AI Engineer at Bosch Global Software Technologies (2023-07 – Present)

• Built an asynchronous FastAPI backend for an Agentic RAG-based AI Assistant handling internal community data, deployed on AWS Fargate and actively used by 500+ users in production.
• Implemented JWT-based authentication, RAG data siloing, and hybrid retrieval pipelines combining vector and keyword search, query enhancement, and self-reflection, improving answer quality by 50% and reducing hallucinations by 40%.
• Implemented LiteLLM-based model routing and fallback strategies to optimize LLM utilization, reducing costs by 30% and improving response latency by 35%, while leveraging the RAGAS framework for evaluation, testing, and performance benchmarking.
• Designed and developed an AI-driven engineering simulation platform that transforms natural language requirements into automated simulation workflows, reducing manual configuration effort by 70%.
• Implemented intelligent knowledge retrieval, planning, execution orchestration, and validation workflows to automate simulation analysis and generate engineering recommendations with real-time progress tracking.
• Developed an AI-based tool to compare versions of large engineering specification documents using semantic similarity, cutting manual review time by approximately 50% compared to line-by-line comparison.
• Extracted requirements from PDF and Word documents, matched related requirements using embeddings, and classified changes as Added, Modified, Deleted, or Unchanged, achieving 98% accuracy with a false-positive rate of approximately 2%.
• Generated structured change reports and summaries; adopted by 20+ engineering teams, reducing manual effort in specification review by 35%.
• Built an Agentic AI system that crawls & generates blogs on emerging technologies in the automotive industry (EVs, ADAS, autonomous driving, connected vehicles).
• Integrated Tavily and Google Scholar tools for autonomous research.
• Implemented an orchestrated multi-agent workflow with weekly automation.

Bachelor of Computer Science Engineering in Computer Science Engineering – KLE Technological University (formerly B.V.B. College of Engineering and Technology) (2019-2023)