Job Description
What you’ll build
- Multi-agent development pipelines that pick up a task, write and review code, spin up preview deployments, capture results, and hand back to humans for approval (we have a live POC — you’ll harden and extend it).
- Agents that plug into our commerce stack — Magento, Shopify, Node.js services — to automate merchandising, QA, content, and ops.
- AI-driven marketing automation across paid, SEO, and lifecycle.
- The orchestration and observability that makes agents run reliably at scale — queues, workers, retries, evals, monitoring.
Who you are
- You build things nobody asked you to build. Your GitHub, demos, or shipped products speak louder than your résumé.
- You get obsessed with hard problems and don’t put them down until they work.
- You operate with extreme ownership — you take a vague goal and return a working system.
- You ship, measure, and iterate fast. You treat shipping as a craft.
- You were already building with LLMs and agents before anyone made you — because you wanted to.
- You want to build the thing, not just talk about it.
Must-haves
- Strong, current coding ability in Python and/or Node.js / TypeScript.
- Hands-on experience building with LLM APIs (Claude, OpenAI, etc.) and agent orchestration (LangGraph, CrewAI, AutoGen, MCP, or hand-rolled).
- At least one real, demonstrable agentic/LLM project you can show and walk through (link required).
- Production deployment experience (Vercel, Railway, AWS, or similar) — not just notebooks.
- Comfort with the glue: APIs, webhooks, queues, vector DBs, auth.
- Strong written and spoken English.
- Willing to travel to our Ahmedabad office periodically (flights + stay covered) to build closely with the core team.
Nice to have
- eCommerce stack exposure (Magento, Shopify, Hydrogen).
- RAG, evals, and agent/prompt observability.
- Full-stack / front-end for internal tooling.
- Open-source contributions, hackathon wins, or a build-in-public presence.
