Senior Software Engineer, Embedded Vision Systems (Toronto)
Summary
Senior engineer on a small team at a VC-backed agtech company, owning the embedded real-time computer-vision pipeline: multi-camera capture, edge model inference under power and thermal limits, camera/sensor drivers, and reliable updates to a remote, intermittently online fleet. Core stack: C/C++/Rust, embedded Linux, vision models on edge devices.
Senior Software Engineer, Embedded Vision Systems
Remote within Canada (Toronto-based engineering team) · Full-time, permanent
About the company
Our client is a VC-backed agtech company using real-time computer vision and AI to help growers produce more fruit and vegetables on existing farmland, more sustainably. Its system runs in the field on a single embedded compute module, works offline first, and ships to a remote fleet that is only online some of the time.
The role
As a senior engineer on a small, multidisciplinary team, you will work across the embedded, cloud and mobile parts of the stack, with the embedded vision system as your core focus.
What you will do
- Own the capture and inference pipeline, from multi-camera video through the models to storage, with no dropped frames.
- Run demanding vision models in real time within fixed power and thermal limits.
- Bring up cameras and sensors at the driver level.
- Turn sensor data into accurate per-tree geolocation, validated against surveyed ground truth.
- Make field issues reproducible at a desk through replays, synthetic sources, on-device CI and regression tests.
- Build self-diagnostics for frame drops, latency, thermal and power headroom, sensor health and storage.
- Take the next-generation hardware platform from definition to shipped product.
- Ship reliably to a remote, intermittently connected fleet.
Must have
- Strong systems programming in C, C++ or Rust, including where application code meets drivers and hardware.
- Real-time streaming or vision pipeline experience: buffer management, latency, backpressure and zero-copy memory.
- Hands-on experience deploying vision models to edge devices.
- Low-level camera and sensor work: image sensors, capture drivers, timing and synchronization, I2C/SPI/GPIO.
- Comfort with embedded Linux, including kernel work, containers and cross-compilation.
- Experience making embedded or distributed systems observable with metrics, logs and dashboards.
- Self-direction in a small, distributed team.
Nice to have
- ROS or GStreamer, including custom elements.
- Embedded vision processors and neural network accelerators, with profiling experience.
- Kalman filtering, multi-object tracking, geometric state estimation or photogrammetry.
- Telemetry pipelines, time-series metrics and automated reporting.
- A background in agriculture, robotics, autonomy or a similar field.
Compensation and benefits
$135K–$175K CAD base, plus a competitive benefits package, team events and retreats, and real influence as an early employee.
