Headline

AI Engineer & Team Lead · Agentic AI Platforms | LLMs · RAG · LangChain · PydanticAI · Langfuse | AWS · Kubernetes | BMW TechWorks

About

I'm an AI Engineer with 6.8 years of experience in software engineering — the last 2–3 years spent almost entirely on designing and shipping AI systems and LLM-based applications in production. My foundation is in building end-to-end backend systems — APIs, distributed services, and cloud-native infrastructure on AWS and Kubernetes. That engineering depth is what lets me build AI systems that aren't just clever prototypes, but things that actually hold up at scale. At BMW TechWorks, I lead the development of a framework-agnostic Agents SDK — part of a larger internal AI Platform that gives ML engineers and data scientists shared infrastructure to build and run their workloads. I proposed the architecture, aligned stakeholders, and built the critical pieces hands-on: the middleware layer that maps user-defined agent logic to LangChain or PydanticAI at runtime, multi-session memory, stateful agent execution, tool orchestration, and a complete RAG pipeline using MongoDB. Most recently, I integrated full observability into the SDK using Langfuse — including self-hosted deployment on our own infrastructure. At SAP, I introduced Generative AI into production applications — conversational assistants and AI-driven automation workflows that improved operational efficiency by 40%. I also led platform extensibility work that opened the application to third-party developers, driving a 30% increase in usage. Earlier in my career, I worked on applied ML in healthcare — building a computer vision diagnostic system using OpenCV, Keras, and RetinaNet that analysed medical images and generated clinical reports. That's where my instinct for making ML actually work in the real world came from. I focus on solving one key problem: taking AI systems from prototype → scalable, production-grade deployments. 𝗜 𝘀𝗽𝗲𝗰𝗶𝗮𝗹𝗶𝘇𝗲 𝗶𝗻: 🔹 LLM Systems — RAG, Agentic Workflows, Memory Systems, Tool Orchestration, Prompt Engineering 🔹 Generative AI — LangChain, PydanticAI, Vector Databases, Embeddings, Langfuse 🔹 Cloud & Architecture — AWS, Kubernetes, Microservices 🔹 Backend Engineering — Python, FastAPI, APIs, Distributed Systems, System Design 🔹 Observability & DevOps — CI/CD, Prometheus, Grafana, Langfuse

Location

Pune District

AI stack

langchain

langfuse

langgraph

mcp

LiteLLM

vLLM

Work experience

Senior Software Development Engineer ( AI / Cloud) at BMW TechWorks India, Jul 2025 - Present · 1 yr 2 mos, Pune District, • Leading the Agents offering within an internal AI Platform that provides ML engineers and data scientists shared infrastructure to build, test, and deploy ML workloads at scale. • Proposed, architected, and bootstrapped the Agents SDK from scratch — presented the initial design to stakeholders and drove alignment on technical direction before writing the first line of code. • Designed the framework-agnostic middleware layer that dynamically maps user-defined agent logic to LangChain or PydanticAI at runtime — decoupling business logic from framework internals entirely. • Built the full RAG pipeline end to end — document ingestion, embedding generation, vector search, and context injection — using MongoDB as the vector store, making retrieval-augmented agents a native SDK capability. • Implemented multi-session memory and stateful agent execution, enabling agents to maintain context across sessions without the developer managing persistence manually. • Architected and integrated end-to-end observability for the SDK using Langfuse — including self-hosted deployment on internal infrastructure, evaluating tradeoffs between ClickHouse Cloud and self-hosting, and designing the full user onboarding and offboarding journey. • Mentoring 3 junior developers, setting technical standards, conducting design and code reviews, and making sure the systems we ship hold up under real production conditions.

Developer at SAP Labs India, Oct 2023 - Jul 2025 · 1 yr 10 mos, Bengaluru, • Worked on Generative AI use cases including conversational chatbots and AI agents for workflow automation, improving operational efficiency by 40% • Led extensibility development for a cloud-based platform, enabling partners and third-party developers to extend application capabilities via Cloud Application Programming, driving a 30% increase in overall application usage • Designed and built cloud-native applications using React, Node.js, and microservices architecture on AWS (Lambda, API Gateway, DynamoDB) • Improved backend performance by 40% and system scalability by 50% through optimized API and database design • Implemented secure authentication (OAuth 2.0) and enhanced system reliability • Integrated monitoring (Prometheus, Grafana) and CI/CD pipelines to ensure reliable production deployments

Developer Associate at SAP Labs India, Dec 2021 - Oct 2023 · 1 yr 11 mos, Bengaluru, Karnataka, India

Software Engineer at GS Lab, May 2021 - Dec 2021 · 8 mos, Pune Division, • Led a small team to build a computer vision-based ML system for medical diagnostics using OpenCV, Keras, and RetinaNet • Designed image processing and classification pipelines for detecting and analyzing biomarker indicators • Developed automated report generation translating ML outputs into structured clinical insights • Worked on applied machine learning in a real-world healthcare use case, building strong foundations in model training and inference

Software Engineer at Bosch Engineering and Business Solutions, Aug 2020 - May 2021 · 10 mos, Tamil Nadu, India, • Developed full-stack applications using React and Node.js, delivering scalable and responsive user interfaces • Owned end-to-end feature development across frontend and backend systems • Built strong foundations in distributed systems, APIs, and scalable application architecture

Associate Software Engineer at Bosch Engineering and Business Solutions, Aug 2019 - Aug 2020 · 1 yr 1 mo, Coimbatore, Tamil Nadu, India

Project Trainee at Bosch Engineering and Business Solutions, Jan 2019 - Jun 2019 · 6 mos, Tamil Nadu, India, • Independently designed and built a blockchain-based smart contract system using Hyperledger Fabric and Go to automate and secure transactions across business entities • Eliminated contract tampering and transformed a manual approval process into a fully automated, transparent, and auditable workflow

Education

Bachelor of Technology - BTech, Information Technology, Kalinga Institute of Industrial Technology, Bhubaneswar