Headline

AI Product Engineer | Ex-Atlan | Open to new roles — global / remote / freelance | About me: aravindan.cc

About

I build AI that holds up in production. Most AI demos impress in a room and break in the real world. I'm the person who takes them from "look, it demos" to "look, it's reliable" — the eval loops, the tool boundaries, the boring plumbing that makes agents actually usable. Over the past 2+ years, I've shipped multi-agent systems in production: - Built an AI support agent (Nora) that deflected 75 tickets/month — up from 18 — projecting $60K–$100K in annual savings. LangGraph orchestration with Glean RAG and specialist sub-agent routing on Zendesk. - Built Mothership, an autonomous ticket-to-PR pipeline that reads a Linear ticket, spins up a sandbox, clones a repo, implements changes, and opens a review-ready GitHub PR. Lifted first-pass correctness from ~40% to ~75%. - Shipped a Description Agent that auto-generates documentation for tables and columns by learning from lineage and usage patterns. Covered 700K+ assets across 50 customers in 14 days — roughly 450× faster than manual work, saving ~110K hours. - Built a Metrics Glossary Agent that auto-discovers KPIs from BI tools and SQL lineage, reducing metric documentation effort by 70%. My toolkit: LangGraph, Python, TypeScript, RAG, LLM evals, FastAPI, React, Postgres, Docker, Cloudflare. I work across Claude, GPT, and Gemini. I'm currently looking for my next role — specifically, a cross-functional team shipping AI-native products end-to-end. Open to global (with visa), remote, and freelance opportunities. Portfolio: aravindan.cc Github: https://github.com/aravindsriraj

Location

Chennai

AI stack

anthropic apis

claude code

cursor

langgraph

openai

hermesagent

Work experience

Software Engineer II at Atlan, Sep 2025 - Jun 2026 · 10 mos, • Developed an autonomous agent on the deep agents framework to extract KPIs from BI tools and SQL lineage. • Automated documentation workflows, reducing manual effort by 70% and enhancing efficiency. • Created an AI system that auto-generates documentation for 700K+ data assets, achieving a 90%+ user acceptance rate.

Software Engineer I at Atlan, Aug 2024 - Sep 2025 · 1 yr 2 mos, • Architected and deployed end-to-end internal AI applications to streamline engineering workflows. • Developed Mothership, an AI-native system that automates ticket-to-PR transformation, enhancing functional correctness. • Built Nora, a multi-agent support automation platform, improving operational efficiency and saving projected costs.

Software Engineer Intern at Atlan, Feb 2024 - Aug 2024 · 7 mos

Data Science Intern at Scaler, Aug 2022 - Dec 2023 · 1 y 4 mo, India

Education

M.Tech CSE, Business Analytics, Vellore Institute of Technology

Public links

github: https://github.com/aravindsriraj

aravindan.cc: https://aravindan.cc/