Agentic AI Architect, Member of Technical Staff

About the role

Micron's Smart Manufacturing and AI team is looking for an ambitious Member of Technical Staff (MTS) engineer to deliver Agentic AI solutions that support its memory solutions business. The role combines applied Agentic AI engineering with production systems work: architecting autonomous AI agents and their data pipelines, building CI/CD paths to Kubernetes and containers, and profiling and evaluating agent performance.

Key responsibilities

• Design and develop autonomous AI agents with multi-step reasoning, planning and tool execution to automate manufacturing workflows.
• Implement Agentic frameworks that orchestrate LLM interactions with internal APIs, databases and software tools.
• Build and maintain agent evaluation harnesses to benchmark accuracy, reliability and task success.
• Analyze and profile agent and inference workloads (tool-calling latency, multi-agent orchestration) using observability and profiling tools such as PyTorch Profiler, LangSmith and OpenTelemetry.
• Implement multi-agent systems using the Agent-to-Agent (A2A) protocol and the Model Context Protocol (MCP).
• Create and maintain CI/CD pipelines for deploying agents to Kubernetes and containerized cloud environments.
• Design security controls for agents, including sandboxing, access control and mitigation of prompt-injection and tool-misuse risks.
• Mentor engineers on agentic system design and optimization.

Required profile

• Bachelor's or Master's degree in Computer Science, Statistics or a related field; Ph.D. is a plus.
• 8+ years of relevant industry experience, including large-scale data/ML pipelines and/or production Agentic AI systems.
• Experience optimizing inference engines (vLLM, TensorRT-LLM) and building GenAI applications and AI agents.
• Experience building end-to-end systems that test, evaluate and deploy autonomous agents.
• Experience with Kubernetes for deploying and scaling AI agent workloads.

Required skills

• LLMs, prompt engineering, function calling, tool use and Chain-of-Thought reasoning.
• RAG pipelines (document ingestion, chunking, embeddings, vector search).
• Preferred: Microsoft Agent Framework or Google Agent Development Kit, LangChain/LangGraph/CrewAI, TypeScript or Rust.
• Preferred: multi-agent systems, agent swarms, agent profiling and evaluation tooling, computer vision or signal processing.