Product Engineer - AI Agents

About the role

Beam builds self-learning AI agents and an operating system where intelligence grows through autonomous discovery. As a Product Engineer you will build features end to end, decide how they should work and answer for what you ship, often starting from an incomplete problem and getting something in front of users quickly. The work spans traditional software and AI, covering interfaces, APIs, backend services and agent capabilities.

Key responsibilities

• Build product features end to end, from schema to interface
• Turn a loosely defined problem into something users can try quickly
• Build agent capabilities: tool calling, retrieval, state, memory, streaming and guardrails
• Learn how features behave in production and make them more reliable
• Work closely with Product and Design and stay close to customer feedback
• Watch the cost and latency of what you ship
• Take early-stage ideas from nothing to a working MVP

Required profile

• 4+ years building and shipping software that customers use
• Has owned features from problem definition through production
• Comfortable with incomplete requirements and a light process
• Strong fundamentals: testing, debugging and operating what you ship
• Uses AI coding tools daily and answers for what they produce

Required skills

• Full-stack range across backend services, databases, APIs and interfaces
• Working knowledge of LLMs: tool calling, retrieval, context, streaming and how models differ
• Nice to have: TypeScript, React, Next.js, Python, Postgres or cloud infrastructure
• Nice to have: experience at an early-stage startup or high-ownership engineering environment
• Nice to have: agent evaluation, observability or production AI systems; side projects, open-source contributions or technical writing

What we offer

• Competitive compensation and benefits, equity
• Flexible working arrangements and work-from-anywhere
• Parental leave
• A team that values high standards, meaningful work and learning by doing