Agent Engineer

About the opportunity We are seeking engineers who build and operate LLM agents in production and have direct visibility into how those agents are used inside a company. What we are looking for - You have shipped an agent that real users depended on and handled failures in production. - You have designed practical ways to measure whether an agent improved or regressed after a change. - You have operated agents that outlive a single request, including scheduled jobs, long-running work, or cloud sandboxes. - You have seen how internal assistants are adopted, where usage stalls, and what makes teams trust or reject them. - You can discuss tradeoffs in reliability, evaluation, observability, cost, memory, reusable skills, tool integrations, and MCP surfaces. Application process and compensation The first step is a short conversational AI interview. There is no coding exercise or take-home assignment. Strong applicants may be invited to a live 30-minute conversation. Compensation is $100-$500 for the completed conversation, depending on the depth and relevance of experience. Work arrangement - Remote contract opportunity. - Flexible scheduling for the initial application and interview process. - Candidates should bring specific examples and practical lessons from production systems. This opportunity is presented by HumanitApp for an independent hiring partner. HumanitApp is not the employer and does not control selection, contract terms, or payment. Availability and terms may change.