Software Engineer, AI (Early Career)

Ande • New York, NY

1d

About The Position

Ande is hiring an early-career Software Engineer, AI to build the intelligence behind the Ande Agent, a concierge and team of subagents that turn user requests into confirmed, paid, policy-compliant entertainment bookings. The agent handles real transactions for executives and coordinates with real venues. The engineer will ship to production from the first weeks, own parts of the agent end to end, and work closely with senior engineers who provide code review and support.

The work includes improving agent planning, tool use, and recovery; building typed capabilities for searching, holding, booking, paying, and messaging; and creating confirmation proposals for consequential actions. The role also develops evaluation scenarios, graders, and LLM judges; structures venue-conversation data into useful memory and supply facts; and ships typed classifiers for consent, triage, ranking, and urgency with thresholds, timeouts, and fallbacks. The team uses AI coding agents daily and expects the engineer to help improve the guides, skills, and checks used with them.

Requirements

• A degree in Computer Science or Engineering.
• Either one year of professional experience with strong internships, or 1–2 years building with AI.
• Experience building something with LLMs that real people used, such as tool use, multi-step agents, structured outputs, or evaluations; a side project, hackathon project, or internship project qualifies.
• Solid TypeScript or Python skills and strong fundamentals in data structures, APIs, and databases.
• Curiosity about model behavior and the habit of measuring results before and after changes.
• Daily fluency with AI coding agents and the judgment to read and take ownership of their output.
• Clear communication, ability to take feedback, and a bias toward shipping work.

Nice To Haves

• Internship or early-career experience at a startup or AI-first company.
• Experience with evaluations, prompt engineering, or agent frameworks beyond tutorials.
• Experience with search, ranking, or recommendation.
• Experience with React, Node, and Postgres.
• Experience shipping products in hospitality, events, payments, or marketplaces.

Responsibilities

• Improve how the concierge and its sourcing, comparison, and venue-coordination subagents plan, call tools, and recover from failures.
• Investigate failures in real conversations, implement fixes, and verify that the fixes work.
• Build and extend typed agent capabilities for searching, holding, booking, paying, and messaging.
• Develop proposals that users confirm before the agent takes consequential actions.
• Write evaluation scenarios, graders, and LLM judges; add fixed bugs as evaluation cases and help maintain a trustworthy evaluation suite.
• Build systems that turn venue conversations into structured facts to improve future recommendations.
• Ship typed classifiers for consent, triage, ranking, and urgency, with labelled thresholds, timeouts, and fallbacks, progressing from shadow to live rollout.
• Use AI coding agents and help improve the guides, skills, and checks that support their use in the codebase.

What This Job Offers

• Job Type: Full-time
• Career Level: Entry Level