Agent Engineer
Posted 1hrs ago
Employment Information
Report this job
Job expired or something wrong with this job?
Job Description
Agent Engineer assessing production LLM agent reliability, evaluation, observability, and adoption. Mercor pays for expert interviews on real-world agent systems.
Responsibilities:
- Build and operate LLM agents in production
- Assess agent reliability, evaluation, and adoption
- Run scheduled jobs, long-running agent work, and cloud sandboxes
- Analyze how agents are used and adopted inside companies
- Work with internal monoagents connected to company data, shared company memory, reusable skills and playbooks, and tool/MCP surfaces
- Improve visibility into agent actions and costs
- Complete a conversational AI screening interview
- Participate in a live 30-minute team conversation if selected
Requirements:
- Experience building and operating LLM agents in production
- Shipped an agent that real users depended on and handled production failures
- Ability to evaluate whether an agent improved or worsened after a change
- Experience running agents beyond a single request, including scheduled jobs, long-running work, or cloud sandboxes
- Experience observing adoption of internal company assistants
- Understanding of internal monoagents, company data, shared company memory, reusable skills and playbooks, tool and MCP surfaces, agent observability, and cost tracking
- Must complete a short conversational AI interview
- No coding or take-home assignment required
- Must be available for a live 30-minute conversation if selected
Benefits:
- $100 to $500 for the live 30-minute conversation, paid on completion of the call
- Up to $250 for each successful referral through the unique referral link

















