About the role
Forward-Deployed Engineers are the heart of this firm. You will embed inside some of India's most ambitious enterprises, understand how their operations really work, and build the AI systems that run them: agents, voice, document intelligence, and full workflows, shipped into production stacks that millions of people depend on. You will work with frontier models before most of the industry touches them, alongside a pod of senior engineers who move at a pace most companies have forgotten is possible. If you have been waiting for a role where the systems you build actually carry a business, this is it.
Responsibilities:
- Own a deployment end-to-end, from the diagnostic that scopes it to the production system the customer runs long after we leave.
- Build with the frontier: agent frameworks, voice pipelines, and document models on top of our internal platform rails, with the early access that comes from working directly with the labs.
- Set the evaluation bar for your own work: baselines, regression gates, and monitoring that keep systems honest when real money flows through them.
- Work directly with customer engineers and executives, inside their codebase and their constraints, until the metric moves.
- Feed everything you learn back into the platform, so every deployment starts further ahead than the last one.
Requirements:
- Four or more years building production software, with strong fundamentals in Python or TypeScript and the instincts that only come from shipping.
- Real experience with LLM-backed systems that people actually used: agents, RAG, voice or document pipelines, not prototypes that stayed in a notebook.
- Comfort in ambiguous, high-stakes environments: legacy systems, imperfect data, and requirements that sharpen as you go.
- Communication that holds up in a boardroom as well as a code review. You will work with customer executives as much as their engineers.
- Evidence you can move a number: a before, an after, and an honest account of what you did in between.
Nice to have:
- Experience in financial services, lending, or other regulated industries.
- Time at a startup, a consultancy, or a founding team, anywhere you owned outcomes without a playbook.
- Open-source work, public writing, or anything else of yours we can read before we meet.