originate a market from a need no one had named
What proves itBuyers who describe the need in your terms rather than their own prior ones, and at least one other party organising delivery in the category.
One of five philosophies · the verb
Build the thing.
102 capabilities in this school are things you build, carried across 33 classes. They run from run a program someone else wrote and read its error to originate a market from a need no one had named, and every one of them comes with the proof it takes: the artifact you produce before anybody says you can do it.
102things you can learn to build
33classes, each takeable alone
288.9hours recorded in them
835exercises to do yourself
Find your class
Climb the ladder ↓
One class at a time, each at its own price.
A small one from the bench
What can you actually finish?
Tell it the time you really have and what you are building. It answers with one thing you could hand over at the end, the proof that thing takes, and the class that carries it.
24 hours
4 hours a week for 6 weeks
That is enough time to get to:
operate a hosted assistant over uploaded documents
Order 9 · ConcreteWhat proves itA transcript answering a question findable only in the uploaded set, and a second transcript where it declines a question the set does not cover.
Taught in Custom GPTs — about 12 hours of class and exercises.
One rung up: characterise how your program behaves on paths you did not plan for (order 10) in Advanced Interactive Fiction — about 2 hours, which is 1 weeks at 4 hours a week.
The recurring hour
Make & Do
Make & Do is where you build the thing. Not a lecture: Neek runs it as coworking, so you bring what you are making and the hour supplies tech support, a second pair of eyes, and enough company to keep the energy up on the days the build is boring.
Mon-Thu · 9:00 Pacific · led by Neek
Four mornings a week, the same room carries two of the five philosophies.
🔨 Make
Build the thing. Ship software, make art, put hardware together.
🔑 Own
Choose your tools, and know who can take each one away.
So the hour you show up to for a build question is the same hour where somebody knows which of your dependencies can be withdrawn on a Tuesday.
The ladder
5 rungs apart, and every step in between is named
Order 8 is assemble a working computer from parts you chose, and it is done when you hand over a device that boots, its parts list, the total cost, and one line per part on why that one. Order 13 is reconcile agents from different toolchains into one run, done when you hand over one transcript containing agents from at least two frameworks and one self-hosted backend, plus the script chaining its output into a second task. Pick a rung below and you get everything on it, the proof each one takes, and the classes that teach it.
Bottom rung is where you start. Arrow keys climb once a rung has focus.
Rung 1 of 8 3 things you can learn to do at this rung
Order 14 · 5 things you can learn to do
Paradigmatic
You originate the category. The proof is other people: buyers using your words, or somebody outside your organization building to the thing you made.
originate
originate a replacement stack when the economics change
What proves itA cutover record with cost per outcome before and after, the compatibility surface that kept callers unchanged, and no interruption in the traces.
originate a schema that makes invalid business states unrepresentable
What proves itA schema in production, the invalid state someone tried to enter and could not, and the working agreement that was changed rather than accommodated by a nullable column.
originate an interaction paradigm that holds across markets
What proves itThe paradigm running in at least two markets with materially different constraints, and one party outside your organisation building to it.
originate an operating model for managed agent teams
What proves itA written operating model in use by at least two people, with a recorded instance of it catching a failure that the previous arrangement missed.
Taught in Production Agent Engineering
Nothing in that domain sits on this rung. Pick another rung, or widen the filter.
Order 13 · 16 things you can learn to do
Metasystematic
More than one system, reconciled — including systems that disagree, agents from toolchains that were never meant to meet, and rules about who is allowed to know what.
arbitrategovernre-architectreconcilesynthesise
govern a change to a running system you can undo
What proves itA deploy and a rollback against a live system, with the check that told you which version was serving at each moment.
Taught in Keep It Running
govern a swarm by information hierarchy
What proves itA written information policy, a transcript where a mediator withholds, and paired runs with and without the hierarchy scored on the same goal.
Taught in Production Agent Engineering, AI Alignment, Context Engineering, Critical Thinking and Creativity with AI, Intro to Agents, Prompt Engineering
re architect an agent as a stateless reducer
What proves itThe serialised state from a stopped run, and the same agent resuming from that state in a fresh process to the same result.
Taught in RAG & Memory, Context Engineering, Intro to Agents
reconcile a failing retrieval by reranking and rewriting
What proves itA set of questions the first-pass retriever answered wrongly, the reranked or rewritten run, and the before-and-after scores on the same set.
Taught in Advanced Retrieval Augmented Generation, Context Engineering
reconcile a working set with the context window it must fit
What proves itA packing scheme with the token budget per part, a run where the input exceeded the window, and what the scheme dropped rather than overflowing.
Taught in Context Engineering, Prompt Engineering, Using Large Language Models
reconcile agents from different toolchains into one run
What proves itOne transcript containing agents from at least two frameworks and one self-hosted backend, plus the script chaining its output into a second task.
Taught in Production Agent Engineering, Automate Your Email: No Code AI Automation, Critical Thinking and Creativity with AI, Intro to Agents
reconcile an agents run with a tool that failed
What proves itA run with a deliberately broken tool, the agent's recovery path, and the log line that would have told you it happened.
Taught in Agentic SDLC, RAG & Memory, Context Engineering
reconcile several audience models on one surface
What proves itA live surface serving at least three named audiences, the routing rule written down, per-audience outcomes measured apart, and one contradiction found and resolved rather than averaged away.
select among agent architectures
What proves itA written comparison of at least two architectures on the same workload, with the measurements that decided it and the conditions that would reverse the decision.
Taught in Production Agent Engineering, RAG & Memory, Context Engineering, Intro to Agents
synthesise a long work that stays consistent across many generations
What proves itA work long enough to exceed one context, the record you kept of what was established, and a late passage that depends correctly on an early fact.
Taught in Critical Thinking and Creativity with AI, Prompt Engineering
synthesise a memory hierarchy for an agent
What proves itThe tiers with the rule that moves an item between them, and a transcript long enough that something was demoted, showing the agent still answered.
Taught in Production Agent Engineering, RAG & Memory, Context Engineering, Prompt Engineering
synthesise a metalanguage for a problem domain
What proves itThe notation's grammar, a work produced in it that exceeds the context window, and a consistency check a reader runs to catch contradictions.
Taught in Prompt Engineering, Using Large Language Models
synthesise a retrieval over relationships not just similarity
What proves itA question your vector retriever answers wrongly, the graph you built, and the traversal that answers it correctly.
Taught in Production Agent Engineering
synthesise a self extending agent behind a review gate
What proves itA diff the agent authored to its own prompt or toolset, the benchmark run that accepted it, and a logged rejection with the reason recorded.
Taught in Production Agent Engineering, Intro to Agents, Prompt Engineering
synthesise an accountability regime for an unattended agent
What proves itA week of unattended runs with a per-action trace, and a replay of one run reconstructed from the trace alone.
Taught in Production Agent Engineering, Agentic AI Security: Securing What You Build, Agentic SDLC, AI Alignment
synthesise an agent that carries notes across its own runs
What proves itPaired runs of the same multi-step task with the note store kept and cleared, where the cleared run repeats a step the noted run skips, plus the note text the agent wrote and the later turn that cites it.
Taught in Agentic SDLC, RAG & Memory, Context Engineering, Critical Thinking and Creativity with AI, Prompt Engineering
Nothing in that domain sits on this rung. Pick another rung, or widen the filter.
Order 12 · 10 things you can learn to do
Systematic
Parts you chose, assembled into one system, measured as a whole against work you had not seen when you built it.
constructdesignintegrate
construct a multi agent conversation with turn taking
What proves itA transcript where the speaker order differs between two runs of the same task, both terminating, plus the selection rule in source.
Taught in Production Agent Engineering, Context Engineering, Critical Thinking and Creativity with AI
construct a predictive account of a models behaviour
What proves itWritten predictions made before running held-out inputs, the hit rate against them, and the same account re-tested after a model version change with what it got wrong the second time.
Taught in AI Alignment, Using Large Language Models
construct a retrieval system
What proves itA running system, a benchmark of retrieval quality over held-out questions, and citations traceable to source documents.
Taught in Advanced Retrieval Augmented Generation, RAG & Memory, Context Engineering, Intro to Agents, Prompt Engineering
construct an agent that uses tools
What proves itA run log showing tool selection the author did not script, a benchmark over held-out tasks, and a case where it correctly declined to continue.
Taught in Context Engineering, Intro to Agents, Prompt Engineering
design a chunking strategy against measured retrieval
What proves itThe same question set run against two chunking strategies over one corpus, the retrieval scores for each, and the strategy you kept with the reason.
Taught in Advanced Retrieval Augmented Generation, RAG & Memory, Context Engineering, Prompt Engineering
design a packaged assistant over your documents
What proves itA shared assistant used by at least three people who did not build it, its rubric-scored answers, and one answer citing the live API it called.
Taught in RAG & Memory, Context Engineering, Custom GPTs, Prompt Engineering
design a recursive summariser for oversized documents
What proves itA run over a document several times the context window, a fact list written beforehand, and the recall score of the output against it.
Taught in RAG & Memory, Context Engineering
design a system you can afford to keep running
What proves itThe cost curve before and after a change you made for that reason, at two usage levels, and the feature you removed or capped to get it.
Taught in Automate Your Email: No Code AI Automation, Context Engineering, Keep It Running
design an offer for one segment end to end
What proves itA shipped offer, the per-segment outcome measured apart from the rest of the traffic, and the one element you changed when the outcome disagreed.
Taught in Founding Federation: Smallest Sellable Thing, Founding Federation: Price and Ideate (Week 0)
integrate a change into a project you do not control
What proves itA merged change in a repository you do not own, with the review thread that got it there.
Taught in Intro to Open Source Software
Nothing in that domain sits on this rung. Pick another rung, or widen the filter.
Order 11 · 15 things you can learn to do
Formal
You attack your own work and write down what happened. Every proof at this rung has a red state and a green state, and you have to have seen both.
falsifyjustifymeasureverify
falsify a blended voice by looking for its sources
What proves itThe voice written as instructions, a passage produced in it, at least three attempts by other readers to name its influences, and the one device you changed because it gave a source away.
Taught in Critical Thinking and Creativity with AI
falsify a models alignment guardrails
What proves itThe guardrail as stated, the inputs that defeat it, the mitigation, and a re-run where the same inputs no longer succeed.
Taught in AI Alignment, Context Engineering, Prompt Engineering
falsify a prompt with a benchmark
What proves itA benchmark file with graded cases, a run showing a non-zero failure rate, and a written account of which failures are the prompt's fault.
Taught in AI Alignment, Context Engineering, Prompt Engineering
falsify an applications defences against a named attack class
What proves itA working input that reaches the interpreter, the fix, and a regression test that is red before the fix and green after.
Taught in Agentic AI Security: Securing What You Build, AI-Assisted Attacks: What Actually Happened, Blue Team Common Certification, The Dark Arts (Red Team)
hold several audience models on one surface
What proves itA live surface serving at least three named audiences, the routing rule written down, per-audience outcomes counted separately, and at least one located contradiction between two audiences' paths.
Taught in Practical Propaganda, Prompt Engineering
justify a performed version of your work in another medium
What proves itThe written source and the performed version, with the passage you had to change for the new medium and why.
Taught in Critical Thinking and Creativity with AI
justify a story structure that fits what you want to say
What proves itOne premise mapped onto two structures, what each demands, and the one you took with the reason.
Taught in Critical Thinking and Creativity with AI, Intro to Interactive Fiction
justify a vector index for a latency and corpus budget
What proves itTwo index types built over the same vectors, their measured query latency and recall, and the corpus size and latency budget you chose against.
Taught in Advanced Retrieval Augmented Generation, RAG & Memory, Context Engineering
measure whether a fine tune changed behaviour
What proves itThe training set, a held-out set built from different sources, and base-versus-tuned scores on both.
Taught in Advanced Retrieval Augmented Generation, AI Alignment, Context Engineering, Prompt Engineering
verify a program does what you claimed with a test
What proves itA test that is red on the broken version and green on the fixed one, run in both states.
Taught in Agentic SDLC
verify a rubric against independent graders
What proves itTwo rubric versions, the adversarial inputs that forced the revision, and inter-grader agreement measured before and after.
Taught in AI Alignment, Custom GPTs
verify an agents actions with a critic
What proves itA run log containing at least one rejection and the revision that followed, plus a control run with the critic removed.
Taught in Production Agent Engineering, Intro to Agents, Prompt Engineering
verify model written code against its specification
What proves itThe diff, a review note per changed hunk, and a test run that is red on the parent commit and green on the child.
Taught in Agentic AI Security: Securing What You Build, Agentic SDLC, Prompt Engineering
verify synthetic data preserves the property you need
What proves itThe named property, the generated set, and a measurement of that property on both the real and the synthetic data.
Taught in Using Large Language Models
verify you would know your system broke before a user tells you
What proves itA real failure you learned about from your own alerting with the timestamps to prove it, and one you learned about from a user, with what you added so that one cannot happen the same way twice.
Taught in Blue Team Common Certification, Automate Your Email: No Code AI Automation, RAG & Memory, Context Engineering, Keep It Running
Nothing in that domain sits on this rung. Pick another rung, or widen the filter.
Order 10 · 15 things you can learn to do
Abstract
One thing becomes a kind of thing. The same work now takes inputs whose shape you did not see coming, and the proof is the input that broke it before you handled it.
characteriseclassifygeneraliseparameterise
characterise how your program behaves on paths you did not plan for
What proves itThree traversals by deliberately different strategies, the state each ended in, at least one dead end or contradiction you did not know was there, and what you changed because of it.
Taught in Advanced Interactive Fiction, The Interactive Fiction Intensive
characterise what of a generated artifact you can defend as yours
What proves itA generated passage with the defensible parts marked, each traced to your premise, your voice definition or your own history, and the parts you would not sign named as such.
Taught in Critical Thinking and Creativity with AI
classify salvaged power cells by whether they are safe to reuse
What proves itThe tested cells with their measured capacity, the ones you rejected with the reason, and the pack you built running a load.
Taught in Solarpunk Automation
classify where ai can fill a shape and where it must not choose one
What proves itA scene three times longer after AI expansion with its shape intact, and one decision you refused to delegate, with the reason.
Taught in Critical Thinking and Creativity with AI, The Interactive Fiction Intensive, Prompt Engineering
constrain model output to a schema
What proves itA script that calls a model and parses the result with no try/except around the parse, plus a test that fails when the constraint is removed.
Taught in AI Alignment, Automate Your Email: No Code AI Automation, Context Engineering, Intro to Agents, Prompt Engineering
generalise a corpus into a priming representation
What proves itThe representation, its token count against the budget, and a reconstruction scored against the original on questions written first.
generalise a node build into your communitys system
What proves itA deployed node serving a real group, with the failure each part absorbs named and the fallback stated for each. NOTE: the full DESIGN of a community information system would be order 12, and this class does not reach it — the chain computes to 10 and the capability validator says so. Workshop 5 is a showcase of what you built, not an architecture defended against alternatives. The order-11 rung (verifying the system survives a cut you did not choose) is unwritten.
Taught in Solarpunk Automation
generalise a reasoning prompt pattern
What proves itThe pattern applied to three tasks with paired outputs against a plain prompt, plus one task where it made the output worse.
Taught in RAG & Memory, Context Engineering, Critical Thinking and Creativity with AI, Prompt Engineering, Using Large Language Models
generalise a specification an agent can build from
What proves itThe specification, the artifact an agent built from it unattended, and the list of questions it did not have to ask.
Taught in Agentic SDLC, Prompt Engineering
generalise a tool interface for model use
What proves itThe tool definition, a transcript of a model calling it correctly from the description alone, and a transcript of it recovering from a bad call.
Taught in Context Engineering, Critical Thinking and Creativity with AI, Custom GPTs, Intro to Agents, Claude Model Context Protocol, Prompt Engineering
generalise a worked example into something you reuse
What proves itThree runs whose inputs differ in shape rather than only in value, with no edit to the body between them, and the input that broke it before you handled it.
Taught in Advanced Interactive Fiction, The Interactive Fiction Intensive, Intro to Interactive Fiction
parameterise a prompt template
What proves itThe template, three filled instances whose slot values share no words, their outputs, and the slot you had to add after the third one did not fit.
Taught in AI Alignment, RAG & Memory, Context Engineering, Critical Thinking and Creativity with AI, Custom GPTs, Prompt Engineering
parameterise a service so it is reachable by name
What proves itThe service reached by name from a second device, and the record of what the name resolved to before and after the address changed.
Taught in Solarpunk Automation
parameterise an agent persona
What proves itThe persona definition and three transcripts on unrelated tasks where a reader picks out the same agent, including one where it refuses.
Taught in Automate Your Email: No Code AI Automation, Critical Thinking and Creativity with AI, Custom GPTs, The Interactive Fiction Intensive, Intro to Interactive Fiction, Practical Propaganda, Prompt Engineering, Using Large Language Models
parameterise an offering backwards from its floor
What proves itOne offering with its price, the floor it clears, and the number of clients per month the arithmetic requires.
Taught in Founding Federation: Price and Ideate (Week 0)
Nothing in that domain sits on this rung. Pick another rung, or widen the filter.
Order 9 · 32 things you can learn to do
Concrete
You make the artifact. A script, a schema, an interface, a story, an automation that ran unattended three times while you were somewhere else.
configureoperateproducetransform
configure a node to run on power you generate
What proves itA power budget — draw, panel, battery capacity — and the measured hours the node stayed up unplugged.
Taught in Solarpunk Automation
configure an assistants memory and outside connections
What proves itMemory on with one fact it retained across a fresh conversation, one connector enabled with the scopes it was granted written out, and one source you deliberately did not connect with the reason.
Taught in Critical Thinking and Creativity with AI, Custom GPTs, Claude Model Context Protocol
configure an open model endpoint behind a provider interface
What proves itOne client script, unedited, producing comparable output against both the hosted provider and your own endpoint, with only the base URL swapped.
Taught in Production Agent Engineering, Hosting Open-Source Models, Prompt Engineering
configure inference hyperparameters
What proves itThe same prompt run under two settings, with the difference in output described against the property you were targeting.
Taught in AI Alignment, Context Engineering
elicit output from a model
What proves itA before-and-after pair of prompts with both outputs, and the written criterion the second satisfies and the first does not.
Taught in RAG & Memory, Context Engineering, Critical Thinking and Creativity with AI, Intro to Agents, Prompt Engineering, Using Large Language Models
express a data shape as a schema
What proves itA schema plus a test file containing at least one payload it accepts and two it rejects.
Taught in RAG & Memory, Context Engineering
operate a hosted assistant over uploaded documents
What proves itA transcript answering a question findable only in the uploaded set, and a second transcript where it declines a question the set does not cover.
Taught in Automate Your Email: No Code AI Automation, Custom GPTs
operate a model as a first pass editor
What proves itA redlined draft showing which of the model's edits you took and which you refused, with the standard each decision was made against.
Taught in Critical Thinking and Creativity with AI, Prompt Engineering, Using Large Language Models
operate a no code automation that calls a model
What proves itA live automation with a run history of at least three unattended runs, and the records it wrote at the destination.
Taught in AI Workflows (no code), Automate Your Email: No Code AI Automation, Critical Thinking and Creativity with AI
operate a recorded decision so a later branch depends on it
What proves itA value written at one point and read at a point far enough away that the two are not visible together, two runs differing only in that decision, and the place the difference becomes visible.
Taught in Advanced Interactive Fiction, The Interactive Fiction Intensive
operate a single board computer as an always on service
What proves itThe service reachable after an unplanned restart, with the boot configuration that made it come back written down.
Taught in make-and-do, Solarpunk Automation
operate concurrent strands that surface at the right time
What proves itTwo strands running independently, the condition that decides when each surfaces, and one run where changing that condition changed the order the reader saw them in.
Taught in Advanced Interactive Fiction, The Interactive Fiction Intensive
operate named states instead of magic numbers
What proves itThe named state set, the code that moves between states, and one transition the naming made you notice that a number would have hidden.
Taught in Advanced Interactive Fiction, The Interactive Fiction Intensive
operate state so a branching story neither explodes nor rails
What proves itOne story shown to fail both ways and then fixed with state, with the variable that does the work named.
Taught in Advanced Interactive Fiction, The Interactive Fiction Intensive, Intro to Interactive Fiction
produce a budget for work whose timeline you have observed
What proves itThe completion gate answered out loud, and either a budget built on an observed timeline or a scoped first run with a defined end and a showable output.
Taught in Data Activism: Turn Data Into Power, Founding Federation: Build With Your Heart First, Founding Federation: Price and Ideate (Week 0)
produce a function whose result depends on program state
What proves itThe function, two call sites that do not know about each other, and two runs of the same call producing different results with the state that caused each difference named.
Taught in Advanced Interactive Fiction, The Interactive Fiction Intensive
produce a parts list you can actually source
What proves itThe list with a source and a price against each line, the substitution you made when something was unavailable or too expensive, and the part that arrived wrong.
Taught in Build a Cyberdeck, make-and-do, Solarpunk Automation
produce a playable story from a running skeleton
What proves itYour own dungeon, playable end to end, holding at least one piece of state that a later scene reads back.
Taught in The Interactive Fiction Intensive, Intro to Interactive Fiction
produce a refusal boundary that fires on cases you did not list
What proves itThree declines on cases absent from the boundary text, three accepts on near neighbours you did want, and the near miss that made you rewrite it.
Taught in Context Engineering, Prompt Engineering
produce a reusable system prompt
What proves itThe prompt in daily use, plus two transcripts from different sessions where a preference it encodes shows up without being asked for.
Taught in Automate Your Email: No Code AI Automation, Context Engineering, Critical Thinking and Creativity with AI, Intro to Agents, Using Large Language Models
produce a rubric a model can apply
What proves itThe rubric plus two independent gradings of the same three outputs, with the scores side by side.
Taught in AI Alignment, Critical Thinking and Creativity with AI, Custom GPTs, Prompt Engineering
produce a stopping condition an agent can check on itself
What proves itA run that halts on its own with the condition met, a run that halts with it unmet and says so, and the case where it stopped too early.
Taught in RAG & Memory, Context Engineering, Intro to Agents
produce a usable interface for an agent
What proves itA reachable URL and a recording of a first-time user finishing the task with no spoken instruction from you.
Taught in Claude Model Context Protocol
produce a working script from a worked example
What proves itThe original example, your script, and the run that produced the output you wanted rather than the example's.
Taught in The Dark Arts (Red Team)
produce an offer page that answers a cold message
What proves itA live page, the outbound message that points at it, and the single action it asks for.
Taught in Founding Federation: Smallest Sellable Thing
produce combinations that clash on purpose
What proves itTen pairs made on paper, the two you could build something on, and a written reason each of those two generates where the others only startle.
Taught in Critical Thinking and Creativity with AI, Practical Propaganda
publish a story to a public shelf and get it played
What proves itA story live on The Spaces Between's public shelf, with mood tags and metadata set, and a playtest exchange in both directions.
Taught in Advanced Interactive Fiction, The Interactive Fiction Intensive, Intro to Interactive Fiction
recover a working state from your own version history
What proves itThe history showing a break, the recovery, and the change you kept across it.
Taught in Intro to Open Source Software
reproduce your environment on a machine that is not yours
What proves itThe declaration file, and the same program running on a second machine with no steps taken that are not in the file.
Taught in Build a Cyberdeck, Hosting Open-Source Models, Solarpunk Automation
transform a corpus into an embedded index
What proves itThe index, ten queries with their top results, and your relevance call on each — including the ones the index got wrong.
Taught in Advanced Retrieval Augmented Generation, RAG & Memory, Context Engineering, Intro to Agents
transform a linear story into chained diamonds
What proves itThree diamonds chained, with the trunk demonstrably reading differently depending on state carried through them.
Taught in The Interactive Fiction Intensive, Intro to Interactive Fiction
transform a transcript into a structured record
What proves itThe source text, the field list you wrote first, and the filled record, with every field traceable to a span of the source.
Taught in Automate Your Email: No Code AI Automation, Prompt Engineering
Nothing in that domain sits on this rung. Pick another rung, or widen the filter.
Order 8 · 6 things you can learn to do
Primary
You do the round trip by hand, with nothing helpful in between. A computer becomes parts you chose. A tool call becomes four messages you can name.
executelocaterun
assemble a working computer from parts you chose
What proves itA device that boots, its parts list, the total cost, and one line per part on why that one.
Taught in Build a Cyberdeck
execute a tool call round trip by hand
What proves itA transcript of all four messages — request, tool call, tool result, final answer — from a script with no agent library in it.
Taught in RAG & Memory, Context Engineering, Intro to Agents, Claude Model Context Protocol, Prompt Engineering
locate the line an error points to
What proves itA stack trace, the line you navigated to from it, and one case where the line named was not where the mistake actually was.
Taught in make-and-do
repurpose a retired device into a working server
What proves itAn old phone or single-board computer running Linux and serving something over the local network, named by model, with what it replaced.
Taught in Build a Cyberdeck, Consumer Device Rescue and Defense, Solarpunk Automation
run a model analysis against a hand check
What proves itThe dataset, the model's analysis, and your own computation of one figure alongside it, agreeing or with the discrepancy explained.
Taught in Data Activism: Turn Data Into Power
run an open model locally
What proves itThe server log showing the model loaded, and a client transcript from a separate process receiving the completion.
Taught in Control AI Spending, Hosting Open-Source Models, Intro to Agents, Prompt Engineering
Nothing in that domain sits on this rung. Pick another rung, or widen the filter.
Order 7 · 3 things you can learn to do
Preoperational
Somebody else wrote it. You run it, you find it, you read what comes back — and the proof is that you can say what the error said before you fixed anything.
followlocaterun
follow a model quickstart
What proves itA terminal transcript or notebook showing a successful call you ran yourself, including the model's response.
Taught in Context Engineering, Using Large Language Models
locate a tool in an open tool hub
What proves itA named tool, its argument schema copied from source, and a run showing its behaviour on a deliberately bad input.
Taught in Agentic SDLC, Intro to Agents, Claude Model Context Protocol, Prompt Engineering
run a program someone else wrote and read its error
What proves itA terminal transcript with a real failure, and your written account of what the error said before you fixed anything.
Taught in Agentic SDLC, Build a Cyberdeck, make-and-do
Nothing in that domain sits on this rung. Pick another rung, or widen the filter.
What you can take
The 33 classes that carry Make
Soonest session first, then the rest of the shelf. 288.9 hours of recording across them, 835 exercises, and the companions, workbooks and guided apps that come with the live classes. Every one of them can be taken on its own, at its own price.
The Interactive Fiction Intensive
Live next on Sat 29 AugYou leave able to characterise how your program behaves on paths you did not plan for.
Exercises15
Companion — 7 you get with it
interactive fiction companion
interactive fiction intensive workbook
story reflection journal
vibe authoring toolkit
story structure atlas
ink shape catalog
shape composition guide
about 5.8 hours of material Take it · $200
Automate Your Email: No Code AI Automation
Live next on Fri 4 SepYou leave able to reconcile agents from different toolchains into one run.
Video8 clips, 10.8 hours
Exercises34
Reference — 1 you get with it
automation blueprint
about 21.3 hours of material Take it · $200
Agentic SDLC
Live next on Mon 7 SepYou leave able to reconcile an agents run with a tool that failed.
Video17 clips, 26.5 hours
Exercises46
Companion — 10 you get with it
agentic sdlc companion
agentic sdlc agent memory
agentic sdlc complexity ladder
agentic sdlc compression
agentic sdlc cto mindset
agentic sdlc devops git
agentic sdlc devops ship
agentic sdlc git situations
agentic sdlc outdoor pet
agentic sdlc probabilistic
about 40.0 hours of material Take it · $500
Critical Thinking and Creativity with AI
Live next on Fri 11 SepYou leave able to govern a swarm by information hierarchy.
Video10 clips, 22.4 hours
Exercises21
Companion — 2 you get with it
critical thinking companion
story arc braids
about 29.6 hours of material Take it · $100
Intro to Open Source Software
Live next on Sat 12 SepYou leave able to integrate a change into a project you do not control.
Video2 clips, 3.1 hours
Exercises14
about 8.6 hours of material Take it · $100
Control AI Spending
Live next on Sat 12 SepYou leave able to run an open model locally.
Video2 clips, 3.1 hours
Exercises12
about 8.1 hours of material Take it · $100
Data Activism: Turn Data Into Power
Live next on Mon 14 SepYou leave able to produce a budget for work whose timeline you have observed.
Exercises25
Guided app — 11 you get with it
campaign workbench
data activism companion
method picker
data activism analysis workflow
data activism coalition map
data activism escalation ladder
data activism method selection map
data activism numbers game
data activism rule of one
data activism scoping funnel
data activism three paths
about 8.2 hours of material Take it · $300
Intro to Agents
Live next on Tue 15 SepYou leave able to govern a swarm by information hierarchy.
Video42 clips, 56.7 hours
Exercises32
about 66.7 hours of material Take it · $350
Context Engineering
Live next on Wed 16 SepYou leave able to govern a swarm by information hierarchy.
Video24 clips, 32.7 hours
Exercises282
Guided app — 1 you get with it
context workbench
about 105.2 hours of material Take it · $400
Production Agent Engineering
Live next on Thu 17 SepYou leave able to originate an operating model for managed agent teams.
Video19 clips, 22.3 hours
about 24.3 hours of material Take it · $200
Practical Propaganda
Live next on Sat 19 SepYou leave able to hold several audience models on one surface.
Exercises21
Guided app — 18 you get with it
propaganda
propaganda aplicada dos ecosistemas
propaganda aplicada ecosistema plataformas
propaganda aplicada encuadre ambos lados
propaganda aplicada flujo campana
propaganda aplicada inicio
propaganda aplicada jerarquia combustible
propaganda aplicada mapa perfiles
propaganda aplicada octalysis
propaganda campaign flow
propaganda cross aisle reframing
propaganda fuel hierarchy
propaganda index
propaganda octalysis
propaganda persona stack
propaganda platform ecosystem
propaganda two ecosystems
activist security
about 7.2 hours of material Take it · $100
Hosting Open-Source Models
Live next on Mon 21 SepYou leave able to configure an open model endpoint behind a provider interface.
about 2.0 hours of material Take it · $350
AI Alignment
Live next on Fri 25 SepYou leave able to govern a swarm by information hierarchy.
Exercises1
about 2.2 hours of material Take it · $300
RAG & Memory
Live next on Fri 9 OctYou leave able to re architect an agent as a stateless reducer.
Video9 clips, 11.2 hours
Exercises47
about 24.9 hours of material Take it · $400
Consumer Device Rescue and Defense
Live next on Wed 28 OctYou leave able to repurpose a retired device into a working server.
Reference — 1 you get with it
six roses
about 2.0 hours of material Take it · $100
Advanced Interactive Fiction
You leave able to characterise how your program behaves on paths you did not plan for.
Companion — 2 you get with it
interactive fiction companion
interactive fiction intensive workbook
about 2.0 hours of material Take it
Advanced Retrieval Augmented Generation
You leave able to reconcile a failing retrieval by reranking and rewriting.
Video5 clips, 6.8 hours
Exercises7
about 10.6 hours of material Take it
Agentic AI Security: Securing What You Build
You leave able to synthesise an accountability regime for an unattended agent.
Exercises1
Reference — 2 you get with it
agentic ai security map
agentic ai security frameworks
about 2.2 hours of material Take it
AI Workflows (no code)
You leave able to operate a no code automation that calls a model.
about 2.0 hours of material Take it
AI-Assisted Attacks: What Actually Happened
You leave able to falsify an applications defences against a named attack class.
Reference — 2 you get with it
agentic ai security map
agentic ai security frameworks
about 2.0 hours of material Take it
Blue Team Common Certification
You leave able to falsify an applications defences against a named attack class.
Exercises24
about 8.0 hours of material Take it
Build a Cyberdeck
You leave able to produce a parts list you can actually source.
Companion — 1 you get with it
cyberdeck companion
about 2.0 hours of material Take it
Claude Model Context Protocol
You leave able to generalise a tool interface for model use.
Video3 clips, 5.2 hours
about 7.2 hours of material Take it
Custom GPTs
You leave able to design a packaged assistant over your documents.
Video7 clips, 9.0 hours
Exercises6
about 12.5 hours of material Take it
Founding Federation: Build With Your Heart First
You leave able to produce a budget for work whose timeline you have observed.
about 2.0 hours of material Take it
Founding Federation: Price and Ideate (Week 0)
You leave able to design an offer for one segment end to end.
Exercises34
Guided app — 2 you get with it
pricing bench
social budget
about 10.5 hours of material Take it
Founding Federation: Smallest Sellable Thing
You leave able to design an offer for one segment end to end.
Guided app — 5 you get with it
front page audit
offer designer
outreach
pipeline
value chain bench
about 2.0 hours of material Take it
Intro to Interactive Fiction
You leave able to justify a story structure that fits what you want to say.
Companion — 7 you get with it
interactive fiction companion
interactive fiction intensive workbook
story reflection journal
vibe authoring toolkit
story structure atlas
ink shape catalog
shape composition guide
about 2.0 hours of material Take it
Keep It Running
You leave able to govern a change to a running system you can undo.
about 2.0 hours of material Take it
Prompt Engineering
You leave able to govern a swarm by information hierarchy.
Video58 clips, 63.5 hours
Exercises142
about 101.0 hours of material Take it
Solarpunk Automation
You leave able to classify salvaged power cells by whether they are safe to reuse.
Video4 clips, 3.9 hours
Exercises13
Reference — 1 you get with it
cyberpony
about 9.2 hours of material Take it
The Dark Arts (Red Team)
You leave able to falsify an applications defences against a named attack class.
Video1 clips, 1.0 hours
Exercises3
about 3.8 hours of material Take it
Using Large Language Models
You leave able to reconcile a working set with the context window it must fit.
Video6 clips, 10.7 hours
Exercises55
about 26.4 hours of material Take it
Where Make shows up
9 of the nine paths lean on it
A path is a person you are trying to become; a philosophy is a thing you are trying to do. Same classes, taken in an order somebody worked out for you, with a place to stop.
🧙 The Weird Internet Wizard
Build strange systems that change how people think.
8 classes · next one runs 29 Aug
🤖 The AI BuilderBuild production AI systems from prompt to deployment.
13 classes · next one runs 7 Sep
🧑💼 The FounderFounder with an AI-first company.
9 classes · next one runs 26 Aug
💼 The AI ExecutiveReplace repetitive business operations with AI (a mini-MBA in AI).
6 classes · next one runs 26 Aug
🎨 The AI CreativeA creative studio with AI.
5 classes · next one runs 29 Aug
📊 The AI CitizenA liberal-arts curriculum rebuilt around AI.
4 classes · next one runs 29 Aug
🔓 Digital IndependenceTechnological sovereignty — not just saving money.
8 classes · next one runs 12 Sep
🌱 The Resilience TrackCommunity resilience and continuity.
5 classes · next one runs 12 Sep
🛡️ The DefenderDefend your devices, your identity, your comms and your organisation.
10 classes · next one runs 26 Aug
If one class turns into a habit
Take a whole path instead
Nothing on this page needs more than the single class you came for. If the shelf starts looking better than the title, every path sells itself two ways, monthly, and each one prices its own.
The recordings
Every recording and the written curriculum for a path, watched on your own hours rather than at the times above.
Across the classes on this page that is 288.9 filmed hours and 835 exercises to work through. Anything you finish stays yours.
See the nine paths → The room as wellA seat in every class that path runs this month, live — with the recordings and the curriculum too.
Every date at the top of this page sits on some path, and the path that carries it includes the room.
See the nine paths →Who this is not for
Three ways to be in the wrong room
🔍 You want to know whether it is true
If the job is working out what follows from the evidence rather than shipping something, Think is the verb and Research & Activism is the hour. Make gets to verify and falsify at order 11, and it does it to things you built yourself.
See Think →🔑 You want to know who can take it away
Own is the neighbor, and it is the one philosophy you do not have to choose between — Make & Do carries both, in the same room, at the same 9:00. If your actual question is about dependency and withdrawal rather than construction, start there and come to the same hour.
See Own →🧭 You want somebody to tell you what to build
The hour is coworking: you bring the thing. If you would rather be handed the project, take a class instead — they come with the exercises, the sequence and the artifact to produce, and you can do one at a time without joining anything.
See every class →Start on the rung you are actually on
Bring the thing you are stuck on
Not a plan for a thing. The thing. Find your rung on the ladder, read the proof it takes, take the one class that carries it, and go and produce that proof. Make & Do is Mon-Thu at 9:00 Pacific while you do it — coworking rather than instruction, so you build and somebody unblocks you.