Full Stack Engineer, AI systems

Role overview Build full stack product features for AI systems that can carry out useful, multi-step tasks over time, using persistent context and external tools. The work spans user-facing interfaces, backend services, and AI integrations, with a strong focus on making agent workflows responsive, dependable, and able to recover from failures.

Responsibilities - Deliver end-to-end product features across frontend, backend, and AI integrations. - Design agent workflows for planning, tool use, multi-step execution, and recovery when something goes wrong. - Integrate language models, memory, and external tools into production systems. - Develop real-time interactions with streaming or partial results while managing latency. - Improve reliability through observability, fallback behavior, evaluation, and iteration based on real usage. - Work with machine learning, backend, and product colleagues to take features from idea through release.

Requirements - Strong full stack engineering experience across frontend and backend. - Sound understanding of system design and API architecture. - Experience with language models, retrieval-augmented generation, or AI-powered applications. - Practical judgment in ambiguous, fast-changing work, with ownership of features through production.

Nice to have - Experience with Next.js, Python, Node.js, PyTorch, SQL or NoSQL databases, Docker, or Kubernetes. - Familiarity with OpenAI, Anthropic, or open-source language models.

Benefits and work setup - Work with a small, hands-on team that makes decisions collaboratively and values independent execution. - The interview process is expected to involve three or four conversations, held virtually and/or onsite.