AI Agent Developer

Role overview Build practical AI agents and related workflows for client projects, from clarifying the requirements through implementation, testing, and handover. The work spans model prompts, tools, data retrieval, and safety controls; choosing plain code or an AI approach that fits the task is part of the job.

Responsibilities - Review project specifications, identify gaps, and agree in writing on the agent’s scope and success criteria before building. - Plan and implement agents or AI-enabled workflows, integrating model APIs with REST services, databases, automation tools, or MCP servers as needed. - Add document retrieval when required, along with safeguards such as human approval for risky actions, logging, and cost or rate limits. - Build tests from realistic examples and evaluate results as the system changes. - Document and hand over completed work, explain limitations clearly, and provide agreed post-delivery support for issues within the original scope. - Protect client data, prompts, and credentials.

Requirements - Python and experience building APIs with FastAPI or Django. - At least one LLM agent or assistant deployed for real users. - Familiarity with tool calling, structured outputs, and prompt design. - Experience with REST APIs, webhooks, and JSON. - Experience with n8n or a comparable workflow automation tool. - Strong written English, careful requirement gathering, independent work habits, and reliability in meeting deadlines.

Nice to have - Experience with Claude, OpenAI, or Gemini APIs; LangGraph or the OpenAI Agents SDK; retrieval-augmented generation; and MCP servers.

Benefits and work setup - Remote, project-based contract work. Fees are agreed in writing for each project or milestone; there is no fixed monthly salary or guaranteed workload. - Post-delivery support lasts for the period agreed per project. Fixes and small in-scope changes are covered during that period; new work is estimated separately.