Micron Technology is looking to solve complex challenges in semiconductor component, packaging, and systems engineering by developing data-driven and physics-based Advanced Modeling and AI solutions. The goal is to enhance predictive modeling, intelligent automation, and decision-making across Micron's memory and storage portfolio, ultimately accelerating digital twin development and optimizing fab technology.
Requirements
- Proficient in Python, with over 5 years of experience working with a variety of semiconductor design & process datasets.
- Proficient in leveraging enterprise data platforms such as Snowflake, BigQuery, MSSQL, Oracle, and AWS Redshift for scalable data processing and analysis.
- Hands-on experience building and managing AI/ML projects involving LLMs, RAG, and agentic workflows deployed in semiconductor environments, demonstrated expertise in ML frameworks such as PyTorch or TensorFlow.
- Strong knowledge of domain-adapted LLM training with practical experience, including pretraining, post training on in-domain synthetic datasets, and model quantization for efficient deployment in semiconductor environments.
- Must have experience working with cloud platforms such as GCP, AWS, or Azure, including deployment of ML pipelines in production environments.
- Familiarity with LLM evaluation and benchmarking techniques, including RLHF, prompt tuning, and reward modeling for hardware-aware use cases.
- Experience with CI/CD pipelines and MLOps practices for ML/LLM deployment.
Responsibilities
- Drive the deployment of intelligent, advanced physics or ML-powered modeling solutions that enhance semiconductor process technology development, package development, design, and early manufacturing workflows—leveraging large-scale, unstructured and often sparse data to deliver robust, domain-specific AI systems from problem definition to production.
- Architect and implement agentic AI systems that integrate with Advanced Modeling tools, design automation tools / environments, and product/fab/manufacturing test platforms to accelerate digital twin development for fab technology co-optimization covering wafer, die, and package level models across tech nodes.
- Establish and promote Best Known Methods (BKMs) for deploying LLMs and agentic systems in production environments, ensuring reliability, efficiency, and maintainability.
- Benchmark and evaluate model performance using structured evaluation frameworks, and continuously refine models through prompt tuning, RLHF, and feedback loops.
- Accelerate insight generation from predictive modeling, images, and data.
- Communicate technical insights and solution strategies clearly to both technical and non-technical stakeholders through compelling data storytelling and visualizations.
Other
- Must have a Master’s or PhD in Electrical Engineering, Computer Science, or a related field.
- Excellent communication skills with the ability to collaborate with process technology, silicon design, design verification, validation and product engineering teams, and translate complex AI/ML concepts into actionable solutions for hardware development.
- Proven ability to independently drive AI/ML projects from problem scoping through deployment in semiconductor environments.
- Strong understanding of agentic AI frameworks (e.g., LangGraph, AutoGen) and evaluation tools (e.g., AgentEval).
- Demonstrated ability to translate complex technical concepts into actionable insights for cross-functional teams.
