Edgeble AI
Website:
edgeble.ai
Job details:
What We build
Physical AI breaks after deployment. Models drift, conditions change, and on the edge there is no cloud to catch it. Edgeble self-correcting runtime keeps deployed AI accurate on-device, without stopping inference. It is in production at Tier-1 manufacturers and hardware-agnostic across edge NPUs.
The Role
Own the system layer of the runtime: the full vision-inference substrate from photons to tensors. That means deep work in the camera path, including V4L2, sensor and ISP pipelines, and capture timing, as well as the NPU runtime, including vendor SDKs, scheduling, thermal behavior, and memory behavior under sustained load. The core challenge: in the field, a capture-path or silicon problem looks like a model problem. This role is about making the system layer observable and correctable so the runtime can tell the difference. You will also extend the runtime to new silicon targets. It already runs on multiple NPU platforms; you will add the next ones.
How We Work
Edgeble is agent-native. We build with agentic coding workflows daily on internal platform tooling already set up for it. You direct the agents; your judgment goes on what agents cannot do: correction logic, validation design, and what better means. If you would rather type every line yourself, this role will frustrate you, self-select accordingly.
You
Strong embedded Linux depth with Open Source Mainline kernel/u-boot contribution is required. V4L2/media subsystem experience is central to this role, including sensor bring-up, subdev pipelines, and ISP tuning exposure. NPU or accelerator runtime experience on any vendor stack, such as Rockchip, NXP, Qualcomm, STM, TI, or similar, matters. You should be comfortable reading kernel-side behavior when userspace numbers do not add up. Agent-augmented working evidence matters, as above.
Why Here
Own a defined layer of a production, patent-pending runtime at the moment it scales. Early-team equity, direct work with a founder with 19 years of experience across silicon, embedded systems, Edge AI, Linux kernel/U-Boot maintenance, and physical systems, in the working style most teams are still debating.
Click on Apply to know more.
