Edge Deployment Engineer · SDE-2
Put intelligence where the world is — on the edge, in real time, in the field.
The mission
Awareness begins at the sensor — a live RTSP feed, a GPU the size of your palm, and no room for lag. You'll take deep-learning models and make them run in real time on edge hardware in the field, where a dropped frame has consequences. This is the first, hardest step of the arc: signals → awareness.
The stack you'll live in
What you'll own
- Deploy and optimize computer-vision / deep-learning models on edge hardware — Jetson, embedded GPUs, cameras.
- Build real-time video pipelines with NVIDIA DeepStream / GStreamer; ingest and process RTSP at scale.
- Squeeze out every millisecond: TensorRT, CUDA, quantization, and latency / power budgets.
- Own the edge lifecycle — build, flash, deploy, monitor, update — reliably and in the field.
- Make it robust: systems that run 24/7 in unforgiving conditions.
What we're looking for
- 4+ years shipping CV / ML on edge or embedded devices in production (SDE-2 level).
- Hands-on with NVIDIA DeepStream / GStreamer / TensorRT / CUDA and embedded Linux.
- Strong systems programming: production Python and C++ — Rust a real plus for performance-critical paths.
- A field mindset — it has to work where it's installed, not just on your desk.
Signals that set you apart
- Jetson / edge-GPU optimization, drones, robotics, or industrial deployments.
- Fleet management / over-the-air updates at scale.
Who you are — a Pinacan
You are not simply an employee — you are entrusted with responsibility. You seek truth, earn trust, build for mission-critical environments, and put the mission above recognition.
How we hire for this role
These are the actual rounds — no surprise stages. Posted 30 July 2026. We reply to everyone.
Apply
Check your inbox — we have sent a confirmation with your reference. If it is not there in a few minutes, look in spam once, then write to careers@pinaca.org.