Staff AI Engineer (Taoyuan)
Indexed description
Responsibilities
- Develop, build, and operate production-grade AI agents supporting task automation, decision support, summarization, and human-in-the-loop execution, including the orchestration governing planning, tool use, multi-step state, failure recovery, and escalation to a person.
- Build and maintain the Model Context Protocol (MCP) and enterprise tool layer that exposes Micron's manufacturing systems and enterprise data to agents as well-scoped tools, resources, and prompts, owning tool contracts, schema and version management, and the reviews that keep every exposed capability auditable.
- Ground agents in authoritative, versioned proven experience, working with the manufacturing and engineering experts who own the operational rules to translate them into structured knowledge that agents can reason over reliably as the domain evolves.
- Establish the evaluation, security, observability, and delivery practices required to operate non-deterministic systems in a 24x7 environment: golden datasets and regression testing; tracing, failure-mode review, and production monitoring; prompt-injection defense, permission and data-boundary enforcement, and runtime guardrails; and cost and latency budgets with progressive rollout.
- Deliver enterprise full-stack solutions end to end, spanning C# / ASP.NET and C++ services, RESTful APIs, SQL data models, and Angular front ends on containerized Kubernetes platforms with CI/CD.
- Hold AI components to the same delivery and operational discipline as the MES service stack, from testing and deployment through lifecycle support.
- Define engineering standards and reference implementations, lead architecture and code reviews, mentor engineers on agent and system design, and shape how AI-assisted development and AI-generated output are applied and reviewed within the team.
- Bachelor's degree in Computer Science, Software Engineering, or a related technical field, or equivalent professional experience.
- 8+ years of professional software engineering, including ownership of enterprise systems running in production.
- Strong backend depth in C# / ASP.NET or C++, with SQL schema design, query performance, and transactional integrity.
- Proven experience delivering end-to-end solutions, from RESTful API design and enterprise system integration through deployment with Docker, Kubernetes, and CI/CD.
- Practical front-end development experience with Angular (preferred), React, Vue, or a comparable modern framework.
- Solid command of software design patterns and clean layered architecture, including separation of concerns, dependency inversion, and testable seams.
- Track record of production ownership in environments where failure carries operational consequence, such as manufacturing, MES/MOM systems, logistics, industrial automation, financial transactions, healthcare, ERP, or workflow platforms.
- Experience taking complete AI agent systems from design into production operation, including agent planning, tool contracts, multi-step workflows, state management, and grounding in authoritative data.
- Hands-on experience keeping agents dependable in production: measuring quality, diagnosing failures and hallucinations, defending against prompt injection, enforcing permission boundaries, and ensuring idempotency for side-effecting operations.
- Demonstrated technical influence beyond your own code, such as engineering standards or architecture adopted by others, or engineers you have mentored.
- Ability to learn complex operational domains quickly and work effectively with the people who define them.
- Hands-on MCP (Model Context Protocol) experience, including authoring servers and designing tool, resource, and prompt contracts with attention to version compatibility, permissions, and auditability.
- Experience coordinating multiple specialized agents, including routing and delegation between agents, state handoff across agent boundaries, long-term memory, and escalation paths to human operators.
- Advanced LLMOps practice, including LLM-as-judge evaluation, systematic prompt and model version management, and model selection and routing tradeoffs across a portfolio of agents.
- Knowledge and retrieval engineering, including ontology and schema design for operational domains, chunking and indexing strategy, RAG pipeline architecture, and retrieval quality evaluation.
- Experience establishing Responsible AI, model governance, audit logging, and security review practices for production AI systems.
- Prior work building internal AI platforms, MCP server catalogs, agent skill libraries, or plugin ecosystems consumed by other engineering teams.
- Familiarity with modernizing legacy systems and applying AI to accelerate that work.
To learn more, please visit micron.com/careers
All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, veteran or disability status.
To request assistance with the application process and/or for reasonable accommodations, please contact at [email protected].
Micron Prohibits the use of child labor and complies with all applicable laws, rules, regulations, and other international and industry labor standards.
Micron does not charge candidates any recruitment fees or unlawfully collect any other payment from candidates as consideration for their employment with Micron.
AI alert: Candidates are encouraged to use AI tools to enhance their resume and/or application materials. However, all information provided must be accurate and reflect the candidate's true skills and experiences. Misuse of AI to fabricate or misrepresent qualifications will result in immediate disqualification.
Fraud alert: Micron advises job seekers to be cautious of unsolicited job offers and to verify the authenticity of any communication claiming to be from Micron by checking the official Micron careers website in the About Micron Technology, Inc.
Create a free Caio profile to unlock more results and save your role and location preferences.
Unlock free search