About this role
Summary: Build and improve the intelligence behind the Ande Agent, a concierge system that manages corporate entertainment bookings. Responsibilities include developing tool capabilities, writing evaluations, and shipping production code using TypeScript or Python.
About Ande
Ande makes Corporate Entertainment Effortless. Marketing, sales, EAs and workplace teams use Ande’s AI to book client dinners, sports suites, catering, merch and more. Policies, spend and logistics are managed in one place, with one vendor.
Over 60 enterprises run their entertainment on Ande, with $400 million+ in annual entertainment spend flowing through a network of 93,000 entertainment providers across 90+ cities. Ande has raised $52 million+ in seed and Series A funding from Lightspeed Venture Partners, Redpoint Ventures, Duration Ventures and Sierra Ventures, with participation from Bain Capital Ventures.
About the role
You will build the intelligence behind the Ande Agent. That's the concierge and its team of subagents that take a request from one sentence ("dinner for 12 with the Acme team next Thursday, somewhere private") to a confirmed, paid, on-policy booking.
This is not a chatbot or a demo. The agent acts on real money, on behalf of real executives, with real venues on the other end of the thread. You will ship to production from your first weeks, own real parts of the agent end to end, and work side by side with senior engineers who will review your work closely and help you grow fast.
What you'll work on
- The agent system. Improve how the concierge and its subagents for sourcing, comparison and venue coordination plan, call tools and recover when something goes wrong. Find the failure in real conversations, fix it, and prove the fix.
- Tools the agent can trust. Build and extend the typed capabilities the agent uses to search, hold, book, pay and message, including the proposals a booker confirms before anything consequential happens.
- Evals. Write scenarios, graders and LLM judges, add every bug you fix as a case, and help keep the suite trustworthy enough that changes ship on the number.
- Memory and supply data. Build the pieces that turn every venue conversation into structured facts that make the next recommendation better.
- Fast judgments. Ship typed classifiers for consent, triage, ranking and urgency, each with labelled thresholds, a timeout and a fallback, rolled out from shadow to live.
- How we build. Ship with AI coding agents every day, and help us improve the guides, skills and checks that make them good at our codebase.
What you'll bring
- A degree in Computer Science or Engineering, and either 1 year of professional experience with strong internships, or 1-2 years building with AI.
- Something you built with LLMs that real people used: tool use, multi-step agents, structured outputs or evals. A side project, hackathon win or internship project counts.
- Solid TypeScript or Python, and strong fundamentals in data structures, APIs and databases.
- Curiosity about why a model did what it did, and the habit of measuring before and after a change.
- Daily fluency with AI coding agents, and the judgment to read and own what they write.
- Clear communication and a bias to ship: you take feedback well and turn it around fast.
Even better
- Internships or early roles at a startup or an AI-first company.
- Experience with evals, prompt engineering or agent frameworks beyond the tutorials.
- Experience with search, ranking or recommendation.
- Experience with React, Node and Postgres.
- You've shipped something in hospitality, events, payments or marketplaces.
Equal Employment Opportunity
*Ande is an equal opportunity employer. We are committed to building a diverse and inclusive team and do not discriminate on the basis of race, color, religion, sex, sexual orientation, gender identity or expression, national origin, age, disability, veteran status, or any other characteristic protected by applicable federal, state, or local law. We welcome applications from all qualified candidates and will consider all applicants for employment without regard to any legally protected status*