
Lead AI Engineer
Posted May 7

Posted May 7
This is a fully remote position, open to applicants in Mexico.
• Develop and enhance the evaluation pipeline, focusing on how we assess task completion, identify regressions, and convert production failures into corrected behaviors.
• Manage context effectively: determining what the agent perceives, when it does so, and ensuring coherence in long-term tasks without excessive token usage.
• Create sub-agent patterns: establish when to branch out, how to cleanly compose specialized agents, and maintain the parent agent's control over the final results.
• Oversee document parsing: converting supplier-uploaded invoices, catalogs, and contracts into structured contexts that the agent can analyze.
• Transform Coupa supplier APIs into agent-accessible CLI tools, ensuring clear error surfaces and reasonable defaults for agent users.
• Develop supplier-facing capabilities built on the harness: procedural instructions and tool compositions that enable the agent to manage specific tasks from start to finish.
• Troubleshoot complex production behaviors: investigate why the agent chose a specific path, where confusion arose, and what changes to tools, context, or the harness can resolve it.
• Collaborate with the senior engineer on architectural decisions regarding the harness as you identify gaps while working throughout the stack.
• Possess 3–5 years of experience in delivering production software with strong proficiency in Python (TypeScript is a plus).
• Have successfully launched at least one LLM-powered feature or product, regardless of size, and be able to discuss challenges faced and lessons learned.
• Be a frequent and heavy user of agentic coding tools such as Claude Code, Cursor, Codex, or similar alternatives.
• Have personal projects that demonstrate your passion and creativity, ideally ones that were made feasible by the new generation of AI tools.
• Communicate fluently in English, both in writing and speaking, as skill authoring is detail-oriented and the team operates in English.
• Approach debugging methodically by analyzing logs, traces, and evaluations rather than through speculation.
• Pioneering Technology: At Coupa, we are leading the charge in innovation, utilizing cutting-edge technology to provide our customers with enhanced efficiency and visibility in their spending.
• Collaborative Culture: We prioritize collaboration and teamwork, fostering a culture characterized by transparency, openness, and a collective dedication to excellence.
• Global Impact: Become part of a company where your contributions have a worldwide, measurable influence on our clients, the business, and one another.
Granicus
Omada Health
NineTwoThree Studio
Stride, Inc.
Get handpicked remote jobs straight to your inbox weekly.