Applied ML Engineer, EDA & Physical AI

Trace; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Trace is seeking a senior ML/EDA-focused engineer to advance neural networks and LLM-driven design tools for PCB workflows. You will develop data pipelines, train and deploy models, and integrate model outputs into real-world design environments.

The role emphasizes handling geometry/graph-based optimization challenges, collaborating with PCB and CAD teams, and delivering production-facing tooling at scale. A strong engineering background and practical experience in quick iteration are essential.

Qualifications

• Bachelor’s degree or higher in a quantitative field (ML/EE/CE/CS/Math/Physics).
• Strong background in machine learning and/or electrical engineering.
• 5+ years of relevant industry experience.
• Proficiency in Python and at least one major ML framework (PyTorch or TensorFlow).
• Experience training, fine-tuning, evaluating, or deploying neural networks, LLMs, or similar systems.
• Experience with AWS and/or GCP for training, data pipelines, or deployment.
• Ability to work on structured, geometric, graph-based, or optimization-heavy problems.
• Comfort building production-facing tools, pipelines, and systems.

Responsibilities

• Build and improve neural network and LLM systems for circuit understanding, placement, routing, and design assistance.
• Create training, fine-tuning, and evaluation pipelines for PCB and schematic data.
• Develop model-driven tooling that integrates into production design workflows.
• Collaborate with PCB, CAD, and product teams to encode design rules into model behavior.
• Prototype and productionize systems for retrieval, ranking, generation, scoring, and design-space exploration.
• Build feedback loops between model outputs, design data, and user actions.
• Define and improve metrics for layout quality, constraint satisfaction, manufacturability, and user value.
• Own projects end to end from data prep and experiments to deployment and iteration.

Skills

Python

ML frameworks (PyTorch/TF)

Neural networks

LLMs

Data pipelines

GPU training workflows

Graph/optimization problems

Production tooling

Education

Bachelor’s or higher in ML/EE/CE/CS/Math/Physics

Master’s degree or higher (preferred)

Tools

AWS

GCP

GPU-based training workflows

Model serving infra

Job description

This role sits at the intersection of machine learning, electronic design automation, and physical engineering systems. You will work on neural networks, LLM-powered systems, and data pipelines that improve how PCB designs are created and refined in practice.
What you'll do

• Build and improve neural network and LLM systems for circuit understanding, placement, routing, and design assistance
• Create training, fine-tuning, and evaluation pipelines for PCB and schematic data
• Develop model-driven tooling that integrates into production design workflows
• Work with PCB, CAD, and product teams to encode design rules and constraints into model behavior
• Prototype and productionize systems for retrieval, ranking, generation, scoring, and design-space exploration
• Build feedback loops between model outputs, design data, and user actions
• Define and improve metrics for layout quality, constraint satisfaction, manufacturability, and user value
• Own projects end to end, from data preparation and experiments to deployment and iteration

Required Qualifications

• Bachelor's degree or higher in ML, EE, CE, CS, Mathematics, Physics, or a related quantitative field
• Strong background in machine learning and/or electrical engineering
• 5+ years of relevant industry experience
• Strong experience with Python and at least one major ML framework (PyTorch or TensorFlow)
• Experience training, fine-tuning, evaluating, or deploying neural networks, LLMs, or similar systems
• Experience working with AWS and/or GCP for model training, data pipelines, or deployment
• Ability to work with structured, geometric, graph-based, or optimization-heavy problems
• Comfort working on real-world systems shaped by physical constraints
• Ability to build production-facing tools, pipelines, and systems

Preferred Qualifications

• Master's degree or higher in a relevant field
• Deep familiarity with PCB design, circuit behavior, layout constraints, or manufacturability
• Experience with EDA, CAD, routing, placement, optimization, simulation, or embedded systems
• Experience applying ML to physical systems or hardware-aware problems
• Experience with GPU-based training workflows, distributed training, or model serving infrastructure
• Strong experimental judgment and ability to define useful evaluation metrics
• Comfort in a startup environment with high ownership and fast iteration