Applied AI Engineers

Location: San Francisco, United States; on-site

Company

Builds AI systems and infrastructure that make agent behavior measurable, debuggable, and improvable for high-stakes workflows.

What you will do

• Build systems to aggregate, index, and analyze large-scale, long-running agent interaction data.
• Design post-training and optimization workflows to improve agents internally and for customers.
• Develop agent platform infrastructure, including orchestration, runtimes, and developer tools for defining, testing, deploying, and iterating on agent workflows.
• Build internal tools and infrastructure for experimentation, analysis, and training.
• Integrate agents into customer-facing product workflows and collaborate with external companies and research partners.

Requirements

• Intellectual curiosity, self-direction, and an ability to stay current with research and emerging ideas.
• Clear reasoning about abstract systems and strong depth of thought.
• Ownership of outcomes, responsible experimentation, and business-driven decision-making.
• Strength in at least one area: production data quality and evaluation, real-world agent systems, reinforcement learning and post-training, machine learning fundamentals, infrastructure, data pipelines, model serving, or research-to-product translation.
• Ability to turn ambiguous problems into clear, well-designed plans.

Culture & Benefits

• Research is applied directly to real-world agent data and shipped into the product.
• Significant autonomy and end-to-end ownership of projects.
• Close collaboration with the team, product, customers, external companies, and research partners.
• Work on agent systems supporting finance, legal, operations, and other high-stakes workflows.