Real-Time Computer Vision Architect

What you'll do

• Translate product goals into system, module, and component requirements, including firmware, ISP, and OS constraints.
• Develop and optimize perception algorithms and ISP pipelines for performance, power, and memory on embedded platforms.
• Coordinate sensing tradeoffs and communicate technical requirements with firmware, OS, hardware, and product teams.

What they ask for

• BS in Computer Science, Electrical Engineering, or another relevant field.
• At least 10 years of relevant, proven experience.
• Proven track record developing and improving computer vision perception algorithms and ISP pipelines.
• Experience optimizing computer vision algorithms on embedded platforms and understanding firmware integration, OS, power, memory, and real-time constraints.

Nice to have

• Breadth in computer vision, including classical methods or ML; exposure to vision transformers, foundation models, or vision-language models.
• Experience translating high-level product goals into detailed architectures and system requirements, optimizing ISP pipelines, or leading large multifunctional efforts.

Pay

USD $184,700 to $324,800 base salary

Tools

Computer Vision, ISP, SLAM, PnP-based pose estimation, vision transformers, foundation models, vision-language models, IMUs

AI focus

Develops computer vision perception algorithms; preferred exposure to ML, vision transformers, foundation models, and vision-language models, particularly for resource-constrained environments.