0% climbed0 of 14 camps reached
AI & ML range · Expedition

AI/ML Engineer

Each camp is one subject area. Camps get higher as the subject gets harder. The summit is a job-ready AI/ML engineer.

Elevation profile · 230 h of climbingSummit · Job-ready
Camp reached Test pending Climbing Open Needs permit (paid)
Camp 09 · Phase 3 Modern Deep Learning

Sequence Models & Transformers

Locked · 18 h · Test not taken
TheThecatcatsatsatononitit"it" attends to "cat" (0.8)multi-head attentionadd & layer normfeed-forwardadd & layer normsoftmax(QKᵀ/√d)·V
Self-attention: every token looks at every other token and mixes in what is relevant
This camp needs a permit. Get the full expedition for $129 to open all three packs.
1 · Read pack0%
WEB
Study page
Interactive · diagrams, install guide, glossary
PDF
Study guide
39 pages · PDF
DOC
Editable notes
Word document (.docx)
CHEAT
Cheat sheet
1 page · print-ready PDF
TEST
Area test
20 questions · 70% to pass Not taken
2 · Watch pack0%
Motion-graphics course in production
0 of 2 videos watched · 2 videos · 16 min
Tokens & embeddings
RNNs and their limits
Attention & self-attention
The Transformer block
3 · Build pack0%
Implement attention and a Transformer block
Lab not started · starter kit for Python 3.10+ (pip) or Google Colab
$ python check.py
Starter kit downloadedPending
Bug found from the first failing checkPending
PASS from python check.pyPending
Download the lab kit (.zip)
Hosted lab workspace (42 h included) opens later