Go deeper than any other course will take you.
Kernel engineering, embedded systems and AI compute infrastructure. 115 courses across 16 tracks — live and hands-on, in eight cities or online from anywhere.
Sixteen tracks
From processor architecture to thousand-node HPC. Pick a track to see its modules.
Where to start
The courses teams book first.
KRN-2014 days
Process Lifecycle & Scheduling
How processes are created, scheduled and destroyed, and how scheduling decisions show up as latency in your application.
Practitioner-taught
SAR 10,500Next 8 Nov
DRV-1204 days
Platform Drivers & Device Tree
Describing non-discoverable hardware and writing the drivers that consume that description.
Practitioner-taught
SAR 10,500Next 11 Oct
EMB-1014 days
Yocto Project Fundamentals
BitBake, layers and recipes, from a first build to an image you would actually ship.
Practitioner-taught
SAR 10,500Next 15 Nov
DBG-2204 days
eBPF & bpftrace
Programmable observability: one-liners for immediate answers, custom programs for the questions nothing else answers.
Practitioner-taught
SAR 12,000Next 15 Nov
SEC-2013 days
Multi-Branch Backporting
Taking an upstream fix and applying it correctly across several maintained branches — the core skill of a vendor security team.
Practitioner-taught
SAR 9,000Next 1 Nov
AIC-2004 days
Linux for GPU Systems
What the kernel is doing underneath your training job, and how to tune it. The layer almost nobody teaches.
Practitioner-taught
SAR 10,500Next 15 NovTaught by people who do this work
Every course is delivered by a practising engineer whose day job is the material — with backgrounds across Wind River, Qualcomm, Amazon Robotics, Graphcore and Spire, and kernel work shipping in production today.
That is the whole proposition. Vendor training is taught from a script by professional instructors. This is taught by people who do the job, reading real source on real hardware.
The instructors How we teachTrain your whole team at once
Any course can run privately — on-site at your offices anywhere in the world, or live online for a distributed team. Scheduled around your release calendar, with labs adapted to your hardware.