๐ About Xanadu
Xanadu helps mid-market organisations in the Middle East move from AI experiments to AI that actually runs in production. We’re a Certified Services Partner, Claude Partner Network, building on Claude for the reasoning, workflow and agent layer, and a Canva partner for the creative and communications layer.
We work in pods: a Forward Deployed Engineer onsite with the customer, a Senior AI Engineer leading the build, an AI Data Engineer owning the data and how agents find it, and you shipping it.
โ๏ธ The role
You’re a builder. You take a design and turn it into agents, integrations and workflows that real people use every day, then make sure they keep working.
๐ฏ What you’ll own
- Build. Ship features and agents from the pod’s design.
- Evals and tests. Prove the agent gives the right answer end to end, and prove it still does next week.
- Integrate. Connect agents to the look-up tools the pod’s AI Data Engineer builds, and build the tools agents use to take action in the customer’s systems: updating a CRM record, sending an email, starting a workflow in an ERP or Microsoft 365.
- Deploy and monitor. Get it live, watch it, fix it.
- Document. Runbooks and handover notes a customer can actually use.
- Agent-first. Use Claude Code every day, and help us work out how agents can do more of the build.
- Library. Contribute reusable pieces to our Agent Library.
๐ What good looks like
- 90 days: Claude-certified and shipping inside a live pod.
- 12 months: owning a workstream end to end, and on the track to Senior.
๐ Must-haves
- 2 to 5 years in software engineering.
- At least one LLM-powered thing you built that real users use, at work or a serious side project.
- Python or TypeScript, APIs, git, cloud basics.
- You own your work, you’re curious, and you ask early instead of spinning your wheels.
- Clear written English.
๐ Nice-to-haves
- MCP, agent frameworks, the Claude Agent SDK.
- Data pipelines.
- Front-end skills for internal tools.
- Already Claude-certified.
๐ซ What this isn’t
- Not a demo factory. If it doesn’t run in production, it doesn’t count.
๐งช How we hire
- Intro call (30 min).
- Values and behaviours (45 min).
- Bar-raiser (60 min).
- Working session (90 min): build a small piece of a Claude agent with us, and talk through your choices.
- Decision: within 48 hours of every stage.
๐ Why Xanadu
- Production work, not proofs of concept.
- A Senior who reviews your code properly and wants you to get better.
- Certified on Claude on our dime, and a clear ladder to Senior.
- Equity (ESOP), so you share in what you build.
- No egos, no passengers.
Apply for this role
To apply: send your CV and a link to something you built with an LLM. Tell us what you’d do better next time.
Mention the role: AI Engineer (Bangalore). Questions? careers@xanadu.co.