Headline

Senior AI Engineer @ Fractal | Agentic AI, LLMs & RAG in production | MCP, LangGraph | Ex-CTO

About

The hard part is never the model. It's the messy data, the pipeline that buckles under load, the agent that thinks it succeeded when it didn't, and the 3am page when production breaks. I own all of it — from the first requirements conversation to steady-state support. ⚡ What I do I build agentic AI and LLM systems for enterprise clients at Fractal — Cox, P&G, Fujioka. Multi-agent orchestration (LangGraph), RAG, and MCP servers running live in production, on data pipelines I build end to end. Before that: founding engineer and CTO, taking products 0→1 and building the team. 🛠️ How I work I don't hand off at the model boundary. On any system I'll write the pipeline, wire up the model, design the agents, ship it on Kubernetes, and be the one watching it under real traffic. Python, AWS, Azure, gRPC, MLOps — tools, not identity. New stack or new domain, I ramp fast. The instinct is the same everywhere: understand the problem, ship the smallest thing that works, then make it reliable. 🚀 On my own I build and ship AI products solo, end to end — live on both app stores. That's what a company gets: hand me an ambiguous problem, I come back with something running. Core: agentic AI · LLMs · RAG · MCP · MLOps · speech/ASR · document AI The honest summary is simpler — I'm an engineer who ships, in whatever domain the problem lives.

Location

New Delhi

AI stack

claude code

hugging face

langchain

langgraph

langsmith

AutoGen

Work experience

AI Engineer at Fractal, Apr 2024 - Present · 2 yrs 4 mos, Gurugram, Haryana, India, I have had the opportunity to work on cutting-edge AI projects that drive business transformations through scalable data science and machine learning solutions. My role focuses on the entire machine learning lifecycle, from development to deployment, using a variety of tools and platforms. 🔧 Technologies & Tools: - Kubernetes for container orchestration and scaling ML workloads - MLOps frameworks to streamline and automate machine learning pipelines - AWS & Azure for cloud infrastructure, utilizing services like EC2, S3, ADLS, and Databricks for big data processing - Databricks & Spark for large-scale data transformations and real-time analytics - ADF (Azure Data Factory) for orchestrating ETL processes and managing data workflows - GitHub for version control and collaborative development - Jenkins for continuous integration and delivery (CI/CD) automation

Data Scientist at SPRY, Jan 2023 - Apr 2024 · 1 yr 4 mos, Pune District, Maharashtra, India

Founding Engineer- Full Stack Developer at Ocean Friends, Jan 2022 - Dec 2022 · 1 yr, Remote

Head of Data Science (CTO) at Sarvvid-AI, Jan 2021 - Dec 2021 · 1 yr

Member at House of Geeks Technical Society of IIIT Ranchi, 2019 - 2021 · 2 yrs

Education

Bachelors, Electrical, Electronics and Communications Engineering, Indian Institute of Information Technology Ranchi

Computer Science, Delhi Public School - India