OpenRouter
Applied AI Engineer
About this role
OpenRouter is seeking Applied AI Engineers to build internal agentic workflows that automate support and go-to-market operations. You'll develop end-to-end AI automation tools, integrate with company systems, and focus on reliability and measurable impact rather than flashy prototypes.
What you'll do
- Build and deploy internal agentic workflows from prototype to production on OpenRouter's platform
- Integrate company systems and replace external tools where proprietary solutions make sense
- Establish reliability through evals, guardrails, and monitoring to ensure trustworthy automation
- Measure and demonstrate quantifiable impact on support volume, GTM acceleration, and operational efficiency
- Collaborate with support, GTM, and engineering teams to identify high-leverage workflows to automate
- Ship, measure, and iterate on automation based on real-world usage patterns
What they're looking for
- Agent development and orchestration across multiple frameworks
- Backend systems architecture and end-to-end thinking
- LLM/agentic systems in production environments
- Evaluation frameworks and data-driven validation
- Systems integration and API design
- Security and permission modeling for agent access
- Support or GTM operations domain knowledge
- Software reliability and monitoring practices
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
OpenRouter
OpenRouter builds a platform that routes and manages high-volume requests across language models and AI inference services. The company is hiring software engineers for platform infrastructure and product development, as well as forward-deployed engineers to support customer implementation and scaling.
- Website
- openrouter.io
Likely interview questions
- Walk us through a time you built internal tooling that teams chose over an external solution—what made it win?
- Describe your experience with agent reliability and safety. How do you prevent unintended actions at scale?