Perception Lead

Kovari; San Francisco (CA); On-site; USD 150,000 - 230,000; Full time

Job summary

Kovari is seeking a senior robotics perception engineer to own the perception stack from sensors to actionable representations for both learned policies and classical control. You will work on deployed robots in real hotel environments, addressing variable lighting, glass, obstacles, and repetitive architecture.

You will build perception systems for manipulation and localization, optimize camera pipelines for low latency, and develop real-time detection, segmentation, and pose estimation for

Qualifications

• Expertise in developing low-latency camera pipelines for real-time perception.
• Strong background in classical computer vision and 3D sensing.
• Experience with learned perception methods and state estimation fusion.
• Proven track record building safety-critical perception on deployed robotics.

Responsibilities

• Build perception systems for manipulation tasks and localization/navigation.
• Optimize camera pipelines for low latency and high-frequency performance.
• Develop real-time object detection, segmentation, and pose estimation.
• Address environment-specific challenges: dynamic obstacles, lighting variation, reflective surfaces.
• Debug perception failure modes on deployed hardware in the field.

Skills

Deep C++ & systems

Classical CV: calibration, SLAM, VO

Learned perception: detection, seg., 3

Sensor fusion & state estimation

Safety-critical robotics on deployed h

Job description

The Role

You will own Kovari's perception stack end-to-end—from raw sensor data to actionable representations for both learned policies and classical control. Your systems will run on deployed robots in real hotel environments, handling the messy realities of variable lighting, glass surfaces, temporary obstacles, and repetitive architecture.

What You’ll Do

• Build perception systems for manipulation tasks and localization/navigation
• Optimize camera pipelines for low latency, high-frequency performance suitable for both policy consumption and classical robotics stacks
• Develop real-time object detection, segmentation, and pose estimation for live deployed robots
• Solve environment-specific challenges: dynamic obstacles, lighting variation, repetitive textures, reflective surfaces
• Debug perception failure modes on deployed hardware in the field

What you bring

• Deep C++ and systems knowledge with experience optimizing camera frame pipelines for low-latency, high-hz perception
• Strong background in classical computer vision: extrinsic calibration, SLAM, visual odometry, point cloud registration
• Experience with learned perception: detection, segmentation, depth estimation, pose estimation
• Sensor fusion and state estimation expertise (Kalman filters, factor graphs)
• Track record building safety-critical perception systems on deployed hardware

Values

• The single most important factor for success is pace of learning.
• Refining our craft is something we pursue relentlessly.
• Low ego, high ownership.
• Commitment to the mission. We work in-person, and this isn't a 9-to-5. We're building something hard, and we need people who are all-in.