KERING Applied AI Engineer-Agentic Applications

Product Discovery & Requirements

•       Work with product managers, business analysts and users to understand tasks and pain points. Contribute to product requirements, user stories and acceptance criteria, including how the agent should handle exceptions.

•       Prototype workflows, test assumptions with users and recommend changes to scope or behavior when a simpler approach would better meet the need.

AI Application Development & Evaluation

•       Build agents and workflows in Dify, with Python extensions where needed. Use shared model access and business APIs to implement prompts, tool calls, structured outputs and state management.

•       Develop retrieval and context assembly using shared knowledge services. Improve document preparation and retrieval quality while preserving source references and users’ data access permissions.

•       Integrate agent tools with Java/Spring business services through documented APIs. Implement validation, approval steps and error handling; agree response and streaming contracts with frontend engineers.

•       Maintain scenario-based evaluations and regression tests. Use production traces and user feedback to improve prompts, retrieval, tools and agent behavior, measuring quality, response time and cost.

Delivery Ownership & Collaboration

•       Version code, Dify workflows, prompts, tests and deployment configurations. Work with QA and operations colleagues on releases, production issues and improvements through shared engineering controls.

•       Review partners’ code, prompts, workflow definitions and test evidence. Resolve issues and complete handovers that allow the team to reproduce, support and evolve selected applications.

•       Use AI-assisted development tools in daily engineering work, review their outputs and contribute reusable code and lessons to the team.

Who you are?

Experience & Technical Skills

•       Bachelor’s degree or above in Computer Science, Software Engineering, or a related field.

•       At least four years of software or AI engineering experience, including hands-on delivery and maintenance of LLM applications or AI agents in production.

•       Strong Python skills, including API integration, structured data handling and automated testing.

•       Practical experience with Dify or comparable agent and workflow platforms, including tool use, structured outputs and multi-step execution.

•       Experience with retrieval-augmented generation (RAG), embeddings, vector search and prompt design, and an understanding of how retrieval and context affect application quality.

•       Experience evaluating AI application behavior, reading execution traces and turning failures into tests and improvements.

•       Working knowledge of REST/OpenAPI integration, authentication, access control and databases, together with Git, CI/CD and containerized application delivery.

•       Ability to trace requests across applications, agents and tools, and handle timeouts, retries and user or tenant context correctly.

•       Ability to clarify ambiguous requirements with product and business colleagues and turn them into testable application behavior.

•       luency in Mandarin and good working proficiency in English.

Preferred Experience

•       Experience maintaining or taking over externally developed applications.

•       Experience integrating Python or Dify applications with Java/Spring services, or implementing tools through Model Context Protocol (MCP).

Personal Attributes

•       You ask what users are trying to achieve and contribute ideas about how the application should work.

•       You take responsibility for the software you deliver, including its behavior after launch.

•       You work openly with colleagues, communicate problems early and use feedback to improve your work.