AI Agent Engineer
Job Description
About this role
Agent code is a new genre of software — one where the control flow is statistical, the side effects are real, and the failure modes are exotic. As an AI Agent Engineer for AI training, you will help AI produce agent code that handles tool calls correctly, recovers from errors, and stays within budget on tokens, time, and money.
Key Responsibilities
• Generate and evaluate instruction-response pairs covering agent loops, planners, and reflection patterns.
• Review AI-generated code using LangGraph, CrewAI, AutoGen, or custom agent frameworks.
• Provide feedback on tool-calling design, schema validation, and error recovery.
• Validate AI handling of memory, retrieval, and multi-agent orchestration.
• Evaluate observability patterns (LangSmith, Langfuse, OpenTelemetry) for agent runs.
• Identify subtle issues in retry loops, runaway costs, and unsafe tool exposure.
Ideal Qualifications
• 3• years building production agents or LLM-powered automation systems.
• Deep familiarity with at least one agent framework and one custom-built agent loop.
• Strong grasp of tool-calling APIs across major LLM providers.
• Experience with sandboxed code execution and untrusted-input handling.
• Comfort with Python or TypeScript for agent development.
• Familiarity with browser-control, OS-control, or CRM-integrated agents is a plus.
Project Timeline
• Start Date: Immediate
• Duration: Ongoing
• Commitment: Flexible, 10-25 hours/week
Contract & Payment Terms
• Independent contractor agreement
• Remote work — anywhere in eligible locations
• Weekly payment via Stripe or bank transfer
• Flexible hours
Tune AI to write AI agents that don't burn money — apply now!