Engineering
Forward Deployed AI Engineer
Build AI workflows that read leases, draft tenant replies and reconcile payments — then deploy and tune them in live portfolios.
About the role
Rentesy's AI work is not a chat box on top of a database. It reads leases, drafts tenant replies, reconciles payments and flags maintenance — inside live portfolios where mistakes cost money.
You will sit between product and customers: build the workflow, deploy it with a real team, measure it, and fix what breaks. You will write production code and spend time on calls.
What you'll do
- Design and ship agentic workflows and tool calls on top of Rentesy's data: leases, ledgers, requests and messages.
- Build evaluations and guardrails for outputs that affect money, legal terms or tenant communication.
- Instrument cost, latency and failure modes, and make the boring reliability work visible.
- Work directly with pilot customers to translate messy operations into prompts, tools and product changes.
- Contribute across the stack — TypeScript, Python and SQL — not just prompts.
What you bring
- 2+ years shipping production software, including at least 6 months working with LLM-based systems.
- Practical experience with tool use, structured outputs, retrieval and evals — not only API calls.
- Solid TypeScript or Python, plus enough SQL to investigate data yourself.
- Comfortable talking to customers and debugging in their environment.
- Good judgement about when an LLM is the wrong tool.
Nice to have
- Built voice or multilingual agents for Indian users.
- Worked with Razorpay, UPI or real estate data.
- Experience with monitoring and cost control for LLM products.
Apply
Apply for Forward Deployed AI Engineer
Send your resume and a short answer to one question. We read every application ourselves.
Rentesy on LinkedIn
Follow the build.
We post product launches, open roles and the unglamorous details of building rental software for India. Follow along between applications.