AI Engineer

Role overview

Design and build the AI systems behind features used in existing client products. The work centers on retrieval, prompting, tool use, and evaluation, with a focus on making model behavior measurable and dependable before release and observable in production. You will work with designers and full-stack engineers to align system behavior with the user experience and help clients judge which ideas are ready to build.

Responsibilities

- Design retrieval, prompting, and tool-use pipelines for AI features embedded in client products. - Create evaluation sets and pipelines, and define quality criteria that features must meet before release. - Add guardrails, fallback behavior, and human review steps where incorrect output could have real consequences. - Choose appropriately simple and cost-effective models and approaches, tracking cost and latency from the start. - Monitor production behavior and investigate failures using logs, traces, and real user examples. - Advise clients on the viability of AI ideas, work with designers to represent uncertainty and sources clearly, and document systems for ongoing maintenance.

Requirements

- At least 4 years building production software, including recent hands-on work with LLM applications. - Practical experience with model APIs, prompting, retrieval, and structured outputs. - Experience building evaluation pipelines and measuring AI quality before shipping. - Strong engineering fundamentals in Python or TypeScript. - Ability to collaborate across design and engineering and explain technical tradeoffs to clients.

Benefits and work setup

This is a senior, full-time remote role, with the stated working range of GMT-6 to GMT+6.