AI Agent Engineer

AMD AMD · Semiconductors · Shanghai, China · Engineering

Develops and orchestrates AI agent systems, including multi-agent architectures, MCP servers, and tool-use frameworks, leveraging machine learning models and data analysis to solve team problems. Requires strong Python/C++ skills and experience with agent frameworks.

What you'd actually do

  1. Drive the development of AI agent systems and tools to solve multiple teams' problems (improve efficiency, provide new solutions, etc.), including designing and orchestrating complex multi-agent architectures and implementing machine/deep learning models.
  2. Design, build, and maintain sophisticated multi-agent systems that coordinate autonomous agents for planning, reasoning, tool use, and task execution across diverse workflows.
  3. Develop and integrate MCP (Model Context Protocol) servers, custom skills, and tool-use frameworks to enable context-aware, extensible AI agent capabilities.
  4. Analyze large datasets to extract meaningful insights and guide model and agent development.
  5. Collaborate with cross-functional teams to understand requirements and deliver AI agent-based and ML-based solutions.

Skills

Required

  • Python
  • C/C++
  • Agent frameworks (LangChain, LangGraph, AutoGen, CrewAI, Semantic Kernel)
  • Multi-agent systems
  • Orchestration
  • Tool use
  • RAG
  • MCP (Model Context Protocol) servers
  • Custom skills development
  • Production-grade pipelines
  • Error handling
  • Observability
  • Evaluation/testing frameworks
  • Data analysis
  • Machine learning models

Nice to have

  • Computer architecture
  • Hardware/software co-design
  • Post-silicon system validation
  • Continuous learning
  • Foundational architectures (Transformers)
  • Multi-modal systems
  • MLOps lifecycle
  • Data preparation
  • Versioning
  • Deployment
  • Monitoring
  • PyTorch
  • TensorFlow
  • JAX
  • Publications in top-tier AI/ML conferences

What the JD emphasized

  • Proven experience designing, building, and deploying AI agent systems, including complex multi-agent orchestration
  • Deep hands-on experience with agent frameworks and libraries
  • Strong understanding of agent design patterns: tool use, retrieval-augmented generation (RAG), planning and reasoning loops, memory management, and human-in-the-loop workflows
  • Hands-on experience developing and consuming MCP (Model Context Protocol) servers and custom skills/tools
  • Strong programming skills in Python and C/C++
  • Experience building production-grade agent pipelines with proper error handling, observability, and evaluation/testing frameworks
  • M.S. or Ph.D. in Machine Learning, Deep Learning, Artificial Intelligence, or a related Computer Science field with a focus on AI

Other signals

  • designing and orchestrating complex multi-agent architectures
  • implementing machine/deep learning models
  • develop and integrate MCP (Model Context Protocol) servers, custom skills, and tool-use frameworks
  • Analyze large datasets to extract meaningful insights and guide model and agent development
  • Develop and integrate MCP (Model Context Protocol) servers, custom skills, and tool-use frameworks to enable context-aware, extensible AI agent capabilities