Engineering Intern 1

The Office of the CTO is where innovation takes center stage. We inspire our global technical community to take on grand challenges, understand emerging trends, identify the critical inflections, and drive our sustainability, Environment, Social, and Governance (ESG) practices that will define the next generation of semiconductors and continued impact.

We are looking for an internship candidate who wishes to strengthen their skills in Explainability and Interpretability for probabilistic / Bayesian surrogate models and Bayesian Optimization (BO). The internship will focus on Bayesian surrogate model interpretability and explainable Bayesian Optimization techniques over controlled synthetic functions and real semiconductor data. 

• Explore the Explainable AI literature and come up with relevant methods suitable for interpreting Bayesian surrogate models
• Develop methods to interpret modeled correlations among model outputs and compare learned relationships with known synthetic function / real data properties
• Create post-hoc explanations for the recommended solution selected via BO, clarifying the model evidence and optimization logic that led to the selected recipe
• Implement reproducible experiments on synthetic test functions / real data, performing surrogate modeling, BO and explanation pipelines in Python
• Investigate interpretable representations of correlation, latent structure, and information sharing across outputs
• Develop local post-hoc explanation methods for BO recommended solutions such as feature/parameter attribution, counterfactual analysis, sensitivity analysis, or decomposition of predictive and acquisition quantities
• Define fidelity, stability, consistency, and usability criteria; validate explanations using known synthetic function properties or real data
• Create clear visualizations and concise technical documentation suitable for Data Scientists and process-development stakeholders

• Master or PhD students with strong interest and foundation in statistical machine learning, probabilistic modeling, explainable AI, and Bayesian Optimization
• Strong knowledge on Gaussian Processes, kernel methods, uncertainty quantification, and Bayesian Optimization
• Strong Python programming and debugging skills; experience with PyTorch, GPyTorch, BoTorch, NumPy, Pandas, and visualization libraries is preferred
• Knowledge on correlation analysis, feature attribution / importance, sensitivity analysis, predictive model explanations, and BO decision explanations
• Experience with synthetic benchmark functions, causal reasoning, spatial data, or structured outputs is advantageous
• Self-driven, collaborative, and able to communicate technical results clearly. Semiconductor or process / hardware knowledge is NOT required

We believe it is important for every person to feel valued, included, and empowered to achieve their full potential. By bringing unique individuals and viewpoints together, we achieve extraordinary results.

Lam Research ("Lam" or the "Company") is an equal opportunity employer. Lam is committed to and reaffirms support of equal opportunity in employment and non-discrimination in employment policies, practices and procedures on the basis of race, religious creed, color, national origin, ancestry, physical disability, mental disability, medical condition, genetic information, marital status, sex (including pregnancy, childbirth and related medical conditions), gender, gender identity, gender expression, age, sexual orientation, or military and veteran status or any other category protected by applicable federal, state, or local laws. It is the Company's intention to comply with all applicable laws and regulations. Company policy prohibits unlawful discrimination against applicants or employees.

Lam offers a variety of work location models based on the needs of each role. Our hybrid roles combine the benefits of on-site collaboration with colleagues and the flexibility to work remotely and fall into two categories – On-site Flex and Virtual Flex. ‘On-site Flex’ you’ll work 3+ days per week on-site at a Lam or customer/supplier location, with the opportunity to work remotely for the balance of the week. ‘Virtual Flex’ you’ll work 1-2 days per week on-site at a Lam or customer/supplier location, and remotely the rest of the time.