← All 186 courses
AI-206 AI & Data

Transformers from First Principles

Open the black box: self-attention, multi-head attention, positional encoding, encoder/decoder stacks — implemented small enough to read, then mapped onto production models. Learners can afterwards explain, diagram and defend every arrow in the transformer architecture.

B4Professional
5modules
30skill atoms
1role journey
Curriculum

What this course covers.

5 modules, 30 named skill atoms. Expand any module to see them.

1Attention Mechanics7 skill atoms
Q/K/Vscaled dot-productmaskingsoftmax over scoressqrt(d_k) scalingcausal vs padding masksquadratic sequence cost
2Full Architecture7 skill atoms
multi-headresidualslayer normpositionsfeed-forward blockpre-norm vs post-normRoPE awareness
3Build a Mini-Transformer5 skill atoms
tiny model coded & trained livecharacter-level datasetembedding & unembeddingloss sanity checkssampling from the trained model
4From Paper to Production6 skill atoms
BERT vs GPT familiessize scaling intuitionencoder vs decoder vs encoder-decodermasked LM vs causal LMscaling lawscontext-length growth
5Whiteboard Certification5 skill atoms
diagram-and-defend oral checkpointshape tracing end to endjustify masking choicesdefend head countfailure-mode questioning
Where it fits

AI-206 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 / ML Engineer

Professional stage
AI-205AI-206AI-207

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-206 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