Headline

Data Scientist 1 @ TVS Motor Company | NITW23

About

As a Data Scientist 1 at TVS Motor Company, I contribute to designing and deploying innovative AI solutions, including ServiceMate AI Chatbot using RAG architecture across global regions and a production-scale WIP image validation endpoint leveraging Vision LLMs. My work focuses on optimizing operational reliability, scalability, and response accuracy, while maintaining high standards of performance and quality. With a Bachelor of Technology in Mechanical Engineering from NIT Warangal (2023), I bring expertise in search analysis, large language models, and process validation. I am passionate about applying cutting-edge AI technologies to deliver impactful solutions and streamline complex processes.

Location

Greater Bengaluru Area

AI stack

gemini

hugging face

kiro

langchain

langgraph

mcp

Work experience

Data scientist 1 at TVS Motor Company, Aug 2024 - Present · 2 yrs 2 mos

Data Scientist at TVS Motor Company, Aug 2023 - Jul 2024 · 1 yr, Bengaluru, Karnataka, India, Designed and deployed ServiceMate AI Chatbot using Agentic Retrieval-Augmented Generation (RAG) architecture across 4 global regions, handling 1,000+ daily service queries and improving first-contact resolution and response accuracy. Built a LLMOps monitoring and evaluation framework using MLflow, tracking relevance, groundedness, hallucination rate, and latency, maintaining 90% response quality in production. Built a production-scale WIP image validation endpoint on Azure Databricks using Vision LLMs to automatically detect service activity in workshop images and validate 5S process compliance. Handled 75K+ daily API requests with optimized latency, scalability, and high operational reliability. Engineered a lead identification system with automated outreach (WhatsApp) using DoE optimization, driving 3,500+ incremental monthly walk-ins and 15% increase in conversions. Developed Auto Job Card system using Large Language Models to convert technician voice inputs into structured service complaints, scaling to 3,000+ API requests per day. Implemented observability tracking for latency (<5s), Word Error Rate (<10%), and complaint sufficiency (>85%). Built an automated outbound calling agent workflow using n8n Cloud, scaling to 3,000–4,000 daily calls with retry logic and WhatsApp fallback mechanisms, significantly improving lead engagement and conversion efficiency.

Education

Bachelor of Technology, Mechanical Engineering, National Institute of Technology warangal

National Institute of Technology warangal