Summary:
As a seasoned AI innovator with 3 years and 10 months of experience in harnessing the power of Generative AI and Machine Learning, I have successfully delivered cutting-edge solutions that drive efficiency, innovation, and growth. With a strong background in software development and AI, I have collaborated with cross-functional teams to design and implement scalable AI-driven solutions.

Key Expertise:

• Generative AI: Developed and deployed Large Language Models (LLMs) for various applications, including user story generation, code documentation, and PDF-based question answering.

• Machine Learning & AIops: Implemented AI-driven server health monitoring, anomaly detection, and service ticket classification, leading to significant improvements in operational efficiency.

• Big Data & Data Analysis: Led high-performing teams to develop data-driven solutions using Power BI, Hadoop, Hive, and MySQL.

Achievements:

• Scalable LLM Serving: Designed and deployed a self-hosted LLM inference framework that handled over 1024 concurrent users, scaling from proof-of-concept to production.

• LLM Performance Monitoring: Established real-time monitoring of key LLM performance metrics such as throughput, requests per second, and response latency.

• LLM Observability & Evaluation: Built a system to observe and evaluate LLM performance, improving troubleshooting and model optimization.

• Autonomous Data Analysis: Developed an AI-powered workflow for automated data evaluation, chart generation, code creation, and summarization, accessible via a web interface.

• RAG Pipeline Development: Built a complex pipeline for extracting structured data from Vehicle Components Specification PDFs and applied Retrieval-Augmented Generation (RAG) for applications such as vehicle sales assistance and service center fault analysis.

• User Story Generation using LLMs: Implemented an LLM-based solution that reduced manual effort in writing user stories by 30% on the Jira platform.

• Improved Service Ticket Classification: Enhanced machine learning models for ticket classification, improving accuracy by 15%.

• Optimized Data Replication: Developed an efficient, real-time data replication pipeline from Microsoft SQL Server to ClickHouse, reducing replication time by 50%.

Current Focus:
As a Solution Developer at Tata Technologies, I am currently driving Generative AI innovation for Tata Motors, focusing on scalability, performance optimization, and efficiency gains in AI solutions.

Additional Contributions:

• Implemented AIops for server health monitoring by detecting anomalies in system metrics and visualizing insights on Elastic.co, reducing server downtime by 30%.

• Developed an ML-driven log file/document clustering API using Python, Java, and ElasticSearch, reducing manual document analysis time by 30%.

• Automated Knowledge Transition Planning through a Power BI, PowerApps, and SharePoint dashboard, reducing onboarding time by 40%.

• Spearheaded ETL automation from SQL Server to Snowflake on a real-time basis using Kafka and Python.

• Integrated Continuous Integration & Deployment (CI/CD) pipelines for AI/ML models using Jenkins, reducing manual deployment errors by 80%.

• Managed a high-performing Power BI team for over a year, successfully delivering data-driven insights.

Let's Connect!
I am always open to exploring how AI-powered innovation can drive efficiency and transformation. Feel free to connect with me!