AI Application Engineer (Agents)

Beijing / Shanghai / End-to-end backend project experience / Master’s / 20–37K CNY/month · 200–450K CNY/year / 1 opening each in Beijing and Shanghai

Role overview

Design and develop backend services for agent products, covering agent runtimes, device–cloud coordination and consumer payment flows. Based in Beijing or Shanghai.

Responsibilities

  1. Deliver product features end to end: break down requirements, design technical solutions, implement frontend and backend functionality, and release it.
  2. Build core product capabilities including accounts and permissions, APIs, data management, content presentation and user flows; create reusable components and modules.
  3. Bring agent capabilities to concrete user scenarios by integrating models and tools for multi-turn conversations, streaming output, task execution, progress feedback and error recovery.
  4. Improve results and engineering quality through testing, evaluation and monitoring; optimize task completion, response times, API costs, stability, compatibility and maintainability.
  5. Use AI coding workflows with tools such as Claude Code and Codex for development, debugging and refactoring, with code review and result validation to ensure delivery quality.

Requirements

  1. Master’s degree in computer science, software engineering or a related discipline.
  2. End-to-end backend project experience; able to independently own a service from design to production.
  3. Proficiency in Python and at least one of FastAPI, Django or Flask.
  4. Familiarity with MySQL or PostgreSQL, Redis and message queues, including indexing and SQL optimization.
  5. Familiarity with Linux and Docker, and basic deployment and operations.
  6. Understand HTTP, authentication, caching and concurrency beyond framework usage.
  7. Able to independently develop frontend features with React or Vue.
  8. Good engineering practices including unit tests, Git workflows and technical documentation.

Nice to have

  1. Familiarity with React, React Native, Node.js or related technologies.
  2. LLM application or agent development experience involving tool use, context management, task orchestration or evaluation.
  3. Experience launching products from scratch, serving real users and iterating continuously, including performance, stability or cost challenges.
  4. Experience in any of these areas is preferred: LLM or agent backends with tool calling, RAG or streaming; consumer backends with accounts, subscriptions, notifications or multi-device synchronization; IoT or smart hardware services with device binding, commands or OTA; high-concurrency or multi-tenant SaaS systems.