Perception Engineer

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

Job summary

Kovari is seeking a senior perceptual systems engineer to own our perception stack end to end. You will work on sensor fusion, localization, mapping, and scene understanding for deployed robots in real-world environments.

You will optimize real-time perception on edge hardware, debug field failures, and drive improvements in models and data. This role requires hands-on robotics experience and strong C++/Python skills, with on-site collaboration in San Francisco.

Qualifications

• Strong background in computer vision and 3D geometry.
• Experience shipping perception on real robots or vehicles.
• Experience with SLAM, state estimation, or multi-sensor fusion.
• Proficiency in C++ and Python; PyTorch for learned components.
• Experience with edge inference (TensorRT, ONNX Runtime, or similar).

Responsibilities

• Build sensor fusion across cameras, depth, IMU, and proprioception.
• Own calibration, time synchronization, and sensor health monitoring.
• Develop localization, mapping, and scene understanding for real environments.
• Build detection, segmentation, and 3D representations for manipulation and navigation.
• Optimize perception for real-time inference on edge hardware.
• Debug field failures and drive fixes back into models and data.

Skills

Computer vision

3D geometry

C++

Python

PyTorch

Edge inference

SLAM / multi-sensor fusion

Field robotics

Bias toward action

Tools

TensorRT

ONNX Runtime

Job description

About Kovari
Kovari develops frontier technology to transform physical businesses. We partner with established operators to change how they run, and we build and operate businesses of our own through our subsidiaries.

We are a small team of builders, operators, and researchers. We deploy on site, alongside the businesses we serve, and iterate in the field. We have raised over $6M from leading investors.

We are based in San Francisco and work in person.

The Role
You will own our perception stack end to end, from raw sensor data to representations used by learned policies and classical control. Your systems run on deployed robots and handle the messy realities of the field: variable lighting, glass and reflective surfaces, temporary obstacles, and repetitive architecture.

What You'll Do

• Build sensor fusion across cameras, depth, IMU, and proprioception

• Own calibration, time synchronization, and sensor health monitoring

• Develop localization, mapping, and scene understanding for real environments

• Build detection, segmentation, and 3D representations for manipulation and navigation

• Optimize perception for real-time inference on edge hardware

• Debug field failures and drive fixes back into models and data

What You Bring

• Strong background in computer vision and 3D geometry

• Experience shipping perception on real robots or vehicles

• Experience with SLAM, state estimation, or multi-sensor fusion

• Strong C++ and Python; PyTorch for learned components

• Experience with edge inference (TensorRT, ONNX Runtime, or similar)

• Bonus: publications at CVPR, ICRA, CoRL, RSS, or similar

• Bias toward action: test early, measure, fix what breaks

Values

• Pace of learning trumps everything.

• Relentless refinement of our craft.

• Low ego, high ownership.

• All-in on the mission. We work in person, and this isn't a 9-to-5.