AI Research and Development Engineer - San Jose, CA, United States

Originally posted 27 October 2026 by the employer.

About the role

This role focuses on developing LLM and VLM capabilities and translating research concepts into functional prototypes for semiconductor inspection and complex engineering applications. You will execute model training and adaptation workflows across the entire development lifecycle, from data preparation to deployment and inference optimization.

What you'll do

- Own end-to-end LLM and VLM development workflows, including data curation, preprocessing, model selection, training, fine-tuning, evaluation, and inference optimization. - Design and execute model adaptation strategies such as supervised fine-tuning, instruction tuning, parameter-efficient fine-tuning, preference optimization, and continued pretraining. - Build and maintain training and evaluation datasets, including data cleaning, labeling strategies, synthetic-data generation, quality checks, and contamination controls. - Establish evaluation and benchmarking frameworks covering model quality, task performance, robustness, hallucination, safety, latency, throughput, and cost. - Develop Agentic AI solutions integrating trained or tuned models with tools, planning, memory, and workflow orchestration. - Support the transition of validated AI research and prototypes into reusable platforms and product applications.

What you'll need

- Master’s or PhD degree in Computer Science, Artificial Intelligence, Machine Learning, Electrical Engineering, or a related field (or equivalent practical experience). - Hands-on experience training or fine-tuning LLMs or VLMs end to end, with direct ownership of data preparation, training execution, evaluation, performance optimization, and deployment or inference. - Practical experience with deep-learning frameworks and model-development tooling, specifically Python and PyTorch. - Understanding of transformer-based models, tokenization, optimization, training dynamics, evaluation methods, and common model failure modes. - Experience with distributed training or multi-GPU environments and diagnosing training issues such as instability, overfitting, data-quality problems, and performance regressions. - Demonstrated ability to design experiments and evaluate trade-offs across model architecture, training data, model quality, compute requirements, latency, and cost.

Nice to have

- Experience developing Agentic AI systems, including tool-using agents, planning, memory, multi-agent systems, agent orchestration, and intelligent workflows. - Experience with LLM/VLM adaptation and alignment techniques such as LoRA or QLoRA, supervised fine-tuning, preference optimization, or reinforcement learning. - Experience with LLM/VLM data pipelines, synthetic-data generation, experiment tracking, model registries, distributed training, optimized inference, quantization, or model serving. - Knowledge of semiconductor manufacturing, voltage-contrast inspection, scanning electron microscopy, e-beam inspection, or defect analysis.

Skills: LLM, VLM, PyTorch, wafer inspection, e-beam inspection