About the role

• Lyte is building mapping and perception software for deployment on diverse robot platforms. As a perception engineer, you will develop the core algorithms of our mapping and state-estimation stack. We are looking for engineers that love problems that sit at the intersection of computing, geometry, and optimization.

What you'll do

• Build beautiful maps by fusing multi-modal sensor data (LiDAR, IMU, Camera).
• Capture the world with state-of-the-art 3D reconstruction techniques (SDFs, gaussian splats).
• Craft powerful visualization tools in order to introspect and tune algorithms.
• Profile and optimize code to achieve real-time performance deadlines.

Required Qualifications

• 3+ years of hands-on experience writing C++ and Python.
• Experience implementing and debugging one or more of: Kalman filters, factor graphs, and numerical optimization problems.
• Bundle adjustment, 3D reconstruction, and visual odometry.
• Signed distance functions, neural rendering, and gaussian splats.
• Strong software engineering competency and a penchant for writing well-tested code.

Preferred Qualifications

• Curiosity and a passion for math, physics, or geometry.
• CUDA experience is beneficial.

Benefits (subject to location and local regulations)

• Competitive salary and equity 
• Comprehensive medical, dental, and vision coverage 
• 401(k) retirement plan 
• Flexible vacation and time-off policy 
• Collaborative, fast-paced, and inclusive work environment 
• Opportunity to work on cutting-edge technologies with a highly cross-functional team 

About Lyte

Lyte builds perception systems for Physical AI operating in complex environments. The company combines custom silicon, integrated sensors, and software into a unified platform for robotics, mobility, and next-generation automation. Lyte is headquartered in Sunnyvale, California, with a global presence. For more information, visit www.lyte.ai

If you’re excited about building impactful technology in a dynamic, hands-on environment, we’d love to hear from you!