Navi Mumbai, Maharashtra, India
Summary
Led the development and optimization of AI-powered search tools and RAG-based chatbots, enhancing system performance and user experience for a large user base.
Highlights
Spearheaded the optimization of an AI-Powered Search Tool, reducing search latency to under 100ms while indexing over 90K records and enhancing retrieval relevance by 35% compared to the legacy system.
Engineered and deployed automated data pipelines with multi-API ingestion on AWS EC2, cutting manual data handling by 90% and ensuring hourly data freshness.
Designed and implemented a production-grade RAG chatbot utilizing TypeSense, Qdrant, and Azure OpenAI, supporting 1,300+ daily active users and achieving 82% contextual accuracy and a 98% safe-response rate across 200K+ interactions.
Optimized AI pipelines (RAG & LLMs) using Hugging Face ROBERTa and Cassandra caching, which reduced latency by 21% and infrastructure costs by 28%, enabling scalable deployment for personalized user experiences.
Implemented robust CI/CD workflows for AI services using Docker and GitHub Actions to streamline deployment, and explored large-scale chatbot interaction data to optimize contextual recommendations.
Mentored 3 engineers on GPT-based SQL agents and Stable Diffusion, accelerating feature delivery by 30%.
Led the development of a computer vision module leveraging CLIP, BLIP, and LLaVA, reducing manual annotation hours by 40%.
Recognized as 'Star Performer' in both 2023 and 2024 (Top 2% of engineers) for significant contributions to AI product launches and mentorship.