Backend engineer with 4+ years building distributed systems and data platforms on AWS, now focused on shipping production AI and backend services. I thrive on owning a service from design through production and collaborating directly with engineers and stakeholders to solve real problems.

• 🔭 Currently building LLM-powered conversational assistants and AI-driven data lineage platforms at Amazon
• 🌱 Deep into RAG systems, LangChain, MCP, and production LLM applications
• ⚡ Previously co-developed internal platforms adopted across teams at Barclays
• 💬 Ask me about: System Design, Distributed Systems, Data Pipelines, LLM Architecture

Chennai, India | Dec 2025 – Present

• 🤖 Building an LLM-powered conversational assistant that lets developers query pipeline architecture, lineage, and schema relationships in natural language — cutting debugging and onboarding time
• 🏗️ Architecting Live Data Flow Diagrams (LDFD), an AI-powered platform that auto-generates interactive, end-to-end data lineage diagrams for complex ETL systems
• 🔗 Designing metadata-driven lineage services providing table and column-level lineage and dependency analysis for governance and impact analysis
• 📊 Deployed automated ingestion pipelines moving high-volume telemetry from CloudWatch into Redshift, supporting platforms at scale

Chennai, India | Aug 2022 – Nov 2025

• 🛠️ Co-developed Data Forge Engine (DaFE), a cloud-native, metadata-driven platform letting users configure entire data workflows via JSON — owned end-to-end pipeline generation, scheduling, validation, and notifications
• ☁️ Engineered a cloud-native migration platform moving workloads from on-prem Hadoop/Oracle to AWS, cutting manual migration effort by 60%
• 🔄 Built a fault-tolerant orchestration service for AWS DMS with automated validation and retry handling, removing manual operations and cutting developer effort by 40%
• 📥 Designed a configurable ingestion framework (SCD Type 1/2, full-load, truncate-load) across relational, NoSQL, and file sources using AWS Step Functions and event-driven workflows
• 💰 Built serverless automation (AWS Lambda) for EC2 lifecycle management, reducing infrastructure costs by 40%
• 🚀 Set up GitLab CI/CD pipelines enabling zero-touch releases

Remote | Sep 2020 – Nov 2020

• 🎓 Built full-stack features (backend APIs + frontend) for a responsive School Management System
• 🔒 Developed a secure file-upload validation module blocking malicious uploads

Python PyTorch Milvus Flask Hugging Face LLM

Built a Retrieval-Augmented Generation (RAG) system for Tamil QA using a RoBERTa-Wechsel dense retriever and a fine-tuned XLM-RoBERTa reader, with an automated indexing pipeline continuously ingesting new documents into Milvus. Published fine-tuned models on Hugging Face and exposed inference through a Flask REST API.

Python LangChain LLMs — Barclays GenAI Hackathon — 3rd Place (Regional)

Built a natural-language-to-SQL chatbot using LangChain and LLMs, enabling non-technical users to query production databases through conversational interfaces.

Python Apache Kafka AWS S3/EC2/Glue/Athena

Built a distributed real-time streaming pipeline with fault-tolerant producers/consumers, integrating Glue Data Catalog and Athena for serverless analytics.

Bachelor of Engineering, Computer Science
SSN College of Engineering, Chennai
Aug 2018 – May 2022

• 🎓 Founded "Transfer Learning," a peer group helping non-CS freshers build programming skills
• 🌊 Active volunteer with YRC — beach/lake cleanups and helmet-safety awareness drives