← All 186 courses
AI-219 AI & Data

Edge & On-Device AI

Run models where the data is, under hardware limits that make no allowance for optimism: fixed memory, thermal ceilings, intermittent connectivity and no chance to hot-fix a fleet. Learners shrink a model with quantisation and pruning against an agreed accuracy budget, benchmark it across ONNX Runtime TensorRT Core ML and TFLite, and design an over-the-air update path with rollback. They leave able to argue the latency privacy and cost case for edge over cloud, and to concede when the cloud wins.

B4Professional
5modules
36skill atoms
1role journey
Curriculum

What this course covers.

5 modules, 36 named skill atoms. Expand any module to see them.

1The Case for the Edge7 skill atoms
latency budget without a network round tripoffline and intermittent connectivitydata residency and privacy argumentsbandwidth and cloud inference costdevice fleet heterogeneityhybrid edge-cloud fallback splitwhen the cloud is simply the right answer
2Model Selection for Constrained Hardware7 skill atoms
parameter count against the RAM ceilingMobileNet and EfficientNet familiessmall language models under 3Bdistillation into a student modelaccuracy loss budget agreed upfrontoperator support on the target runtimeMACs versus measured device throughput
3Quantisation and Pruning7 skill atoms
post-training quantisation to int8quantisation-aware trainingcalibration dataset selectionper-channel vs per-tensor scalesstructured vs unstructured pruningper-layer accuracy cliff detectionGGUF and 4-bit weight formats
4Runtimes and Hardware Targets8 skill atoms
ONNX Runtime execution providersTensorRT engine building on JetsonCore ML and the Apple Neural EngineLiteRT (formerly TFLite) with GPU and NPU delegatesNNAPI deprecation and the vendor-delegate pathllama.cpp and ExecuTorch for on-device LLMssilent operator fallback to CPUthermal throttling and sustained throughput
5Fleet Updates and Field Operation7 skill atoms
over-the-air model bundle deliverystaged rollout by device cohortversioned models with rollback triggerssignature verification of model artifactson-device telemetry without raw datasilent accuracy drift in the fieldstorage and download budget per device
Where it fits

AI-219 in a role journey.

This course appears in 1 of our 45 role journeys. Here is what a learner takes immediately before and after it in each.

AI Infrastructure Engineer

Professional stage
AI-211AI-219AI-220

Roles this course serves

The capability ladder

This course is authored to band B4.

Every course we run is written to one rung of the CASI ladder, so a plan can be assembled to take a team from where they are to where they need to be.

What do B1–B6 mean?The CASI Capability Ladder — click to expand

Every course targets a band on the CASI Capability Ladder — our six-band proficiency scale, anchored to open standards (O*NET, ESCO, NICE, NIST AI RMF, Bloom's). A band tells you how deep a course goes, and what evidence proves it.

What the learner can doTypical evidence
B1
AwareUnderstands concepts and vocabulary; uses tools with guidance
Knowledge checks
B2
FoundationPerforms standard tasks correctly in familiar contexts
Guided labs, autograded exercises
B3
PractitionerDelivers complete pieces of work independently
Scenario labs, proctored hands-on exams
B4
ProfessionalHandles production-grade complexity, trade-offs and failure modes
Break-fix drills, design defenses
B5
AdvancedEngineers systems end-to-end under constraints; leads others
Rubric-scored capstones, vivas
B6
ExpertSets direction; recognised authority across teams
Portfolio + panel evaluation

A note on B6. Courses in this catalog target B1–B5. B6 is not taught — it is recognised, through a portfolio and a panel, once someone is setting direction for others. Every journey here is built to land a learner at B5.

Next step

Run AI-219 for your team.

This course runs at several lengths depending on how deep you need to go and how much of it your people already have. Tell us who is being trained and we will scope it.

Add it to a training plan Talk to our team Check your team’s level free