At Skyworks, we work at the intersection of some of industry's most complex RF/analog design and methodology. As part of the central engineering team, you will apply modern AI and ML techniques to semiconductor design and engineering workflows. You will build inference pipelines, curate and engineer design and simulation datasets, and develop applications that use structured reasoning to assist engineers in complex decision-making. This is a software-first role with meaningful domain exposure to semiconductor design and verification.

Responsibilities

• Design and deploy ML inference models for engineering use cases: prediction, classification, anomaly detection, and recommendation over design/simulation data
• Build data engineering pipelines to ingest, clean, label, and version datasets from design and layout databases, as well as simulation and physical verification results
• Develop LLM-based applications with chain-of-thought prompting for guided troubleshooting, documentation Q&A, and methodology assistance
• Implement MLOps practices: model versioning, A/B testing, monitoring for drift, and reproducible training pipelines
• Create APIs and services that integrate ML capabilities into existing EDA and engineering platforms
• Collaborate with domain experts to define features, labels, and evaluation metrics grounded in engineering outcomes
• Stay current with industry-standard techniques in deep learning, transformers, and efficient inference

Required Experience and Skills

• M.Tech/M.S. in Computer Science, EE (with ML focus), or related field from a reputed university
• 3–5 years of professional software development with substantial ML/AI component
• Strong Python skills and experience with ML frameworks (PyTorch, TensorFlow, or scikit-learn)
• Experience building and deploying inference models in production environments (not only notebook prototypes)
• Solid data engineering skills: SQL, ETL pipelines, data versioning, and handling large structured/semi-structured datasets
• Understanding of LLM application development: prompt engineering, RAG architectures, and chain-of-thought reasoning patterns
• Familiarity with model evaluation, cross-validation, and metrics selection appropriate to the problem domain
• Software engineering discipline: testing, code review, Git, and CI/CD for ML projects

Desired Experience and Skills

• Exposure to semiconductor, EDA, or hardware engineering data (SPICE results, DRC violations, parametric test data)
• Experience with MLOps tooling (MLflow, Kubeflow, Weights & Biases) and model serving (TorchServe, Triton, FastAPI)
• Familiarity with vector databases and embedding models for domain-specific RAG
• Understanding of time-series analysis or graph neural networks for netlist/circuit representations
• Publications or demonstrable projects in applied ML (not required, but valued)

#LIPR1