Deep Learning Engineer, Multimodal Vision Platform

Matroid • Palo Alto, CA

6d • Onsite

About The Position

Matroid is a computer vision company whose end-to-end platform enables enterprise customers to train and deploy automated visual inspection across imagery such as EO, IR, X-Ray, CT, and OCT. The Deep Learning Engineer will build the models, capabilities, and systems at the core of Matroid’s multimodal vision platform, taking ideas from research through production to solve visual-inspection problems for customers. The position is based onsite five days a week at Matroid’s downtown Palo Alto office near Stanford and Caltrain.

The role involves analyzing computer vision and deep learning research; designing, adapting, and training models across a broad range of tasks; turning techniques into platform features customers can train and deploy on their own data; and optimizing models and deployment systems for efficient inference on cloud GPUs and edge devices. The engineer will collaborate with product, infrastructure, and field engineering teams to bring capabilities from concept to customer impact.

Requirements

• Master’s degree in Computer Science, Software Engineering, Mathematics, or a related field, or a Bachelor’s degree with equivalent experience.
• At least 3 years of industry and/or academic experience in deep learning, computer vision, and software engineering.
• Experience building ML systems and capabilities, rather than only scripts or one-off experiments.
• Familiarity with modern architectures and techniques, including current foundation models and how they work.
• Strong software engineering skills in Python and PyTorch, with attention to code quality.
• Solid oral, written, presentation, collaboration, and interpersonal communication skills.
• Interest in the full range of the work, from developing video transformer architectures and real-time edge segmentation to making 3D reconstruction work reliably on customer data.

Nice To Haves

• Significant computer vision and deep learning research experience.
• Prior experience applying computer vision to real-world use cases.
• Experience productionizing models, including inference optimization with tools such as TensorRT or ONNX and serving with tools such as Triton Inference Server.
• Experience with 3D reconstruction and scene representation, such as COLMAP, VGGT, 3D Gaussian splatting, or world models.
• Experience building or operating multi-tenant ML products.
• Open-source contributions to ML tooling, personal projects, competitions, or hackathons.
• Previous work at high-growth technology startups.

Responsibilities

• Critically analyze current computer vision and deep learning research and determine which approaches are suitable for the platform.
• Design, adapt, and train models for detection, segmentation, anomaly detection, tracking, video understanding, vision-language, and 3D reconstruction.
• Convert models and techniques into platform capabilities that customers can train and deploy on their own data and use cases.
• Optimize model architectures and deployment systems for fast, cost-efficient inference on cloud GPUs and small edge devices.
• Collaborate with product, infrastructure, and field engineering teams to move models and capabilities from idea to customer impact.

Benefits

• Competitive pay and equity.
• Gym membership reimbursement.
• Free lunch, healthy drinks, and snacks every day.
• Medical, dental, and vision insurance with 100% paid premiums.
• Flexible hours, with room for the occasional day away when life calls for it.
• A budget for hardware or software that will make the employee more effective.
• Regular tech talks on advances in computer vision, deep learning, and engineering.

What This Job Offers

• Job Type: Full-time
• Career Level: Mid Level