Senior Software Engineer, Applied AI

What you’d do

Build and maintain an AI agent harness for long-running work, including planning, tool use, memory and recovery, and integrate OpenAI and Anthropic model APIs. Also build evaluation datasets and replay tests, and measure false positives, missed risks, completion rate and cost. Found on the employer's careers page on Oct 15, 2026. Apply through the link.

You’ll need

• Experience building AI agent harnesses
• Integration with OpenAI and Anthropic APIs
• Model selection by quality, latency and cost
• Evaluation datasets and replay testing
• Prompt injection and unsafe tool use testing