AI Agent Engineer (Forward Deployed / Applied AI)

What You'll Do

1. Build and debug AI Agents in TRAE, owning Prompt design, workflow orchestration, knowledge base integration, Tool Calling, and API connectivity end-to-end.
2. Turn project-level Prompts, workflows, and scripts into reusable Skills, templates, and operational guidelines for the wider team.
3. Stay current on Claude, ChatGPT, and other frontier LLMs — track capability differences, pricing, and limits across providers.
4. Help the team unblock on LLM usage, API integration, Prompt performance, rate-limit handling, and cost optimisation.
5. Design test cases, evaluate Agent output quality, and drive continuous improvement.
6. Work with business users to gather requirements, write technical documentation, and deliver training materials.

What We're Looking For

• Degree in Computer Science, Software Engineering, AI, or a related field; 3–5 years of experience in software development or AI applications.
• Have independently delivered at least one LLM or AI Agent project from start to finish.
• Strong in Python, plus working knowledge of at least one of JavaScript / TypeScript / Java.
• Hands-on experience with Prompt Engineering, RAG, Tool Calling, and knowledge base systems.
• Daily user of AI coding tools such as TRAE, Claude Code, or Cursor.
• Real-world experience with at least two frontier LLMs (e.g., Claude, ChatGPT, Gemini).
• Solid API integration and troubleshooting skills; comfortable reading English technical docs.
• Clear communicator who can partner effectively with non-technical stakeholders.

Nice to Have

• Experience with TRAE custom Agents, Rules, Skills, or MCP servers.
• Hands-on with the Claude API and Claude Code.
• Familiar with platforms or frameworks such as Dify, Coze, or LangChain.
• Delivered projects in enterprise knowledge base, intelligent customer support, or office automation (OA)​.
• Background as an FDE, Solutions Engineer, Technical Support, or Implementation Consultant.
• Portfolio of public AI Agent demos, open-source projects, or technical writing (blog / papers / talks).

What We Offer

• Work on real, shipped AI products — not just POCs.
• Direct exposure to frontier models (Claude, GPT-class) and the latest agent tooling.
• Flat, high-ownership team with fast iteration cycles.