YOUR FOUNDATION-FIRST SYLLABUS
Edge & alternative computing
Explore AI that runs on a nearby device and compare different ways of computing with neural models. Follow the tradeoffs between memory, speed, energy, and useful answers; special hardware is one possible branch.
0 of 101 core ideas demonstrated100 lessons1290 estimated minutes remaining
Already know some of this?
Go straight to an idea’s knowledge check. Passing its different checks without hints carries that evidence into this path. Checking an advanced idea does not award its prerequisites; you can still explore them.
Check what I already knowTHE IDEAS, IN ORDER
Shared credit is automaticBuild your understanding.
Chapter counts use compatible knowledge-check evidence. Reading a lesson does not mark its ideas as demonstrated.
Chapter 1 · Lessons 1–7 Meet an AI system Start with an everyday task, then follow its inputs, model, and output. 7 lessons · 0 of 7 core ideas demonstrated Up next
01
What is AI? Start with an everyday task Common Ground · 11 min
To explore02 Inputs and outputs: what goes in and what comes out Common Ground · 13 min
To explore03 Data and examples: what a computer can learn from Common Ground · 10 min
To explore04 What is a model? A small rule inside a bigger app Common Ground · 10 min
To explore05 Weights: the adjustable numbers in a model Common Ground · 12 min
To explore06 Training and inference: changing a rule or using it Common Ground · 13 min
To explore07 Follow one small AI system from start to finish Common Ground · 9 min
To exploreChapter 2 · Lessons 8–13 How computers represent a problem: How a computer stores a picture or a sentence Explore everyday AI ideas; how AI got here; how computers represent a problem. 6 lessons · 0 of 6 core ideas demonstrated
08
Different ways to solve the same problem Common Ground · 11 min
To explore09 How to tell whether an AI answer is useful Common Ground · 10 min
To explore10 The questions that shaped AI Common Ground · 15 min
To explore11 How a computer stores a picture or a sentence Learning mechanisms · 16 min
To explore12 Describe a computation so someone else can follow it Learning mechanisms · 11 min
To explore13 What a representation makes easy to learn Learning mechanisms · 12 min
To exploreChapter 3 · Lessons 14–19 Inside a neural network Explore learning from useful data; learning to predict; inside a neural network; and supporting ideas. 6 lessons · 0 of 6 core ideas demonstrated
14
From the world to a dataset Learning mechanisms · 11 min
To explore15 The numbers you actually need Math preparation · 10 min
To explore16 A function is a rule Math preparation · 9 min
To explore17 Build a tiny prediction model Learning mechanisms · 11 min
To explore18 Inside an artificial neuron Learning mechanisms · 14 min
To explore19 Why layers need more than multiplication Learning mechanisms · 12 min
To exploreChapter 4 · Lessons 20–26 How computers represent a problem: Trace an example through a batch Explore inside a neural network; how computers represent a problem; math preparation when it is needed. 7 lessons · 0 of 7 core ideas demonstrated
20
Vectors: lists that work together Math preparation · 10 min
To explore21 Read an equation one symbol at a time Math preparation · 10 min
To explore22 Follow shapes through a numerical layer Math preparation · 10 min
To explore23 Trace inputs through layers to a loss Learning mechanisms · 17 min
To explore24 Count work, memory, and parallel limits Math preparation · 11 min
To explore25 Trace an example through a batch Learning mechanisms · 12 min
To explore26 What makes an accelerator useful? Learning mechanisms · 12 min
To exploreChapter 5 · Lessons 27–33 Making training fit the hardware Explore making training fit the hardware; helping a model improve; math preparation when it is needed. 7 lessons · 0 of 8 core ideas demonstrated
27
Why arithmetic counts do not tell the whole speed story Learning mechanisms · 14 min
To explore28 Powers and logarithms, one step at a time Math preparation · 11 min
To explore29 When the computer cannot store the exact number Math preparation · 12 min
To explore30 A derivative is a local change Math preparation · 13 min
To explore31 Several inputs, several rates of change Math preparation · 9 min
To explore32 Choose a direction and a step size Math preparation · 9 min
To explore33 How does a model improve a prediction? Learning mechanisms · 32 min
To exploreChapter 6 · Lessons 34–40 Helping a model improve Explore helping a model improve; making training fit the hardware; math preparation when it is needed. 7 lessons · 0 of 7 core ideas demonstrated
34
Probability without the mystery Math preparation · 10 min
To explore35 Averages over uncertain outcomes Math preparation · 9 min
To explore36 Learn from a sample of examples at each step Learning mechanisms · 11 min
To explore37 How momentum and Adam use earlier gradients Learning mechanisms · 15 min
To explore38 Constrain a solution or change the update Learning mechanisms · 12 min
To explore39 Keep signals and gradients in a usable range Learning mechanisms · 12 min
To explore40 Spend numerical precision where it is needed Learning mechanisms · 13 min
To exploreChapter 7 · Lessons 41–47 Turning language into model inputs: How text becomes tokens Explore turning language into model inputs; following attention through a transformer; preparing and training a language model. 7 lessons · 0 of 7 core ideas demonstrated
41
How text becomes tokens Learning mechanisms · 13 min
To explore42 From token IDs to learned vectors Learning mechanisms · 14 min
To explore43 Attention: queries, keys, and values Learning mechanisms · 13 min
To explore44 From attention scores to a causal mask Learning mechanisms · 17 min
To explore45 How a language model takes its next step Your system, connected · 15 min
To explore46 Follow one token through a transformer Your system, connected · 13 min
To explore47 Inside a language model training step Learning mechanisms · 14 min
To exploreChapter 8 · Lessons 48–54 Learning from useful data Explore learning from useful data; preparing and training a language model; following attention through a transformer; and supporting ideas. 7 lessons · 0 of 7 core ideas demonstrated
48
What can a small experiment tell us? Math preparation · 10 min
To explore49 Clean data without erasing the problem Learning mechanisms · 12 min
To explore50 Protect the examples used to judge a model Learning mechanisms · 15 min
To explore51 Assemble token batches without leaking targets Learning mechanisms · 14 min
To explore52 Choose what a pretraining prediction means Learning mechanisms · 13 min
To explore53 Run several learned comparisons in parallel Learning mechanisms · 10 min
To explore54 The chain rule, one step at a time Math preparation · 13 min
To exploreChapter 9 · Lessons 55–62 Turning language into model inputs: Words, structure, meaning, and context Explore inside a neural network; learning from sequences and sound; following attention through a transformer; and supporting ideas. 8 lessons · 0 of 8 core ideas demonstrated
55
How an error reaches earlier weights Learning mechanisms · 11 min
To explore56 Let the graph carry derivatives Learning mechanisms · 12 min
To explore57 Carry a state from one step to the next Learning mechanisms · 14 min
To explore58 Map one sequence into another Learning mechanisms · 17 min
To explore59 Match information flow to the learning objective Learning mechanisms · 13 min
To explore60 Words, structure, meaning, and context Learning mechanisms · 12 min
To explore61 Represent text with counts and short contexts Learning mechanisms · 14 min
To explore62 Learn the units used by a language model Learning mechanisms · 15 min
To exploreChapter 10 · Lessons 63–70 Preparing and training a language model Explore inside a neural network; preparing and training a language model; learning from useful data; and supporting ideas. 8 lessons · 0 of 8 core ideas demonstrated
63
Help information and learning signals travel Learning mechanisms · 12 min
To explore64 Connect data, gradients, and saved model state Learning mechanisms · 15 min
To explore65 Choose the model that pretraining will fit Learning mechanisms · 15 min
To explore66 Where did this dataset come from? Learning mechanisms · 12 min
To explore67 Build a traceable corpus before spending training compute Learning mechanisms · 19 min
To explore68 Choose how much the model sees of each source Learning mechanisms · 13 min
To explore69 Turn a linear score into a class probability Learning mechanisms · 11 min
To explore70 Control complexity without peeking at the answer Learning mechanisms · 12 min
To exploreChapter 11 · Lessons 71–77 Testing what a model has learned Explore testing what a model has learned; preparing and training a language model; following attention through a transformer; and supporting ideas. 7 lessons · 0 of 7 core ideas demonstrated
71
Will the model work on new examples? Learning mechanisms · 14 min
To explore72 Why a simpler model can predict better Learning mechanisms · 14 min
To explore73 Evaluate the procedure that chose the model Learning mechanisms · 14 min
To explore74 More capacity changes more than one thing Learning mechanisms · 14 min
To explore75 Allocate compute and preserve useful checkpoints Learning mechanisms · 15 min
To explore76 Distances, neighborhoods, and transformations Math preparation · 12 min
To explore77 Give a transformer information about order Learning mechanisms · 12 min
To exploreChapter 12 · Lessons 78–84 Adapting a model after pretraining Explore preparing and training a language model; adapting a model after pretraining; making training fit the hardware; and supporting ideas. 7 lessons · 0 of 7 core ideas demonstrated
78
Continue pretraining without losing sight of the starting model Learning mechanisms · 12 min
To explore79 Track what each adaptation stage changes Learning mechanisms · 15 min
To explore80 Learn response behavior from demonstrations Learning mechanisms · 14 min
To explore81 Learn a scoring signal from response comparisons Learning mechanisms · 14 min
To explore82 Transfer behavior through teacher outputs and selected examples Learning mechanisms · 13 min
To explore83 Account for everything that occupies accelerator memory Learning mechanisms · 14 min
To explore84 Reduce deployment cost while measuring what changes Learning mechanisms · 14 min
To exploreChapter 13 · Lessons 85–92 Exploring other kinds of neural models: Follow voltage, leakage, threshold, and reset Explore using AI to investigate science; exploring other kinds of neural models; how AI got here; and supporting ideas. 8 lessons · 0 of 8 core ideas demonstrated
85
Follow a changing state through time Math preparation · 12 min
To explore86 Match event-driven computation to a scientific workload Learning mechanisms · 14 min
To explore87 Follow voltage, leakage, threshold, and reset Learning mechanisms · 14 min
To explore88 AI grew from several older questions Learning mechanisms · 11 min
To explore89 Update connections from local activity Learning mechanisms · 14 min
To explore90 Learn with spike timing or an approximate backward signal Learning mechanisms · 14 min
To explore91 Use fixed recurrent dynamics and train a readout Learning mechanisms · 14 min
To explore92 Learn a vector field and let a solver define continuous depth Learning mechanisms · 14 min
To exploreChapter 14 · Lessons 93–100 Exploring other kinds of neural models: Recover patterns through energy-based state dynamics Explore generating something new; exploring other kinds of neural models; finding patterns and useful representations; and supporting ideas. 8 lessons · 0 of 8 core ideas demonstrated
93
Predicting a label, generating a possibility Learning mechanisms · 14 min
To explore94 Add up a continuous quantity Math preparation · 9 min
To explore95 Represent probability through invertible maps or energy scores Learning mechanisms · 15 min
To explore96 Recover patterns through energy-based state dynamics Learning mechanisms · 15 min
To explore97 Directions, eigenvectors, and low-rank structure Math preparation · 10 min
To explore98 Compress variation into fewer directions Learning mechanisms · 13 min
To explore99 Learn a dictionary or a neighborhood-organized representation Learning mechanisms · 14 min
To explore100 Compare learning rules, state dynamics, and hardware evidence Learning mechanisms · 14 min
To exploreOptional extensions and comparisons (8)
These lessons deepen or compare the reference design. They do not add requirements to this path’s completion.
How to use this school without getting lostOptional orientation: how paths, side lessons, and practice fit together, plus study habits that make learning stick.Practice, progress, and knowing what you understandOptional orientation: use practice, mistakes, and spaced review to find out what you really understand.Place model computation where memory and latency allow itSplit a model across machines, or squeeze it onto a phone: where to run inference.Plan for traffic, failures, and ongoing resource usePlan capacity and failure handling for a model running on constrained hardware.Save memory by sharing state or recreating itFit training into limited memory by splitting state or recomputing it.Compute attention with less memory trafficFlashAttention-style tiling: the same math with far less memory traffic.Make a large training run recoverable and auditableKeep long training runs recoverable when hardware fails.Load a compatible model and prepare its exact inputGet the weights, tokenizer, and chat template right before running a model on your own hardware.