For people who want to ship the system, not read about it
Getting an answer is easy. Getting it to stay up is the job.
12 classes, first prompt to deployed agent. 230 hours recorded, next live session 9 Nov.
Cancel any time. A class you finish stays yours. Full scholarships exist.
tms path build-ai-systems --stats
- classes
- 12
- recorded
- 230.5 h
- tools
- 48
You write code. Have you built with a language model yet?
Not yet, start with Using Large Language Models, a survey of
the whole field.
Yes, skip it and start at Control AI Spending.
Not sure you can code?
The bill nobody models first
What will this thing cost you to run?
The demo is free. The bill is not. Put in your workload and see it three ways.
A · Pay-per-token API
B · Serverless GPU RunPod, Modal, Replicate
C · Dedicated GPU reserved, or your own box
Fine print you can change
Prefill (reading your input) runs far faster than generation, so the GPU options time them separately. A cached prefix skips prefill on a GPU and bills at the cached rate on an API. Dedicated cards are sized so peak traffic fits. A month is 30.44 days.
Cheapest at 200 tasks a day
,
A · API
,B · Serverless
,C · Dedicated
,Pull a lever ,what each one does to the cheapest bill
The climb
What you walk out able to do
Say how many hours you have and see how far up they reach.
Spend it. Each class costs what it really takes.
- follow a model quickstart
- locate a tool in an open tool hub
- run a program someone else wrote and read its error
- execute a tool call round trip by hand
- locate a claim in its primary source
- locate the devices a passage uses on you
- run an open model locally
- configure an assistant's memory and outside connections
- configure an open model endpoint behind a provider interface
- configure inference hyperparameters
- elicit output from a model
17 more on this rung
- express a data shape as a schema
- operate a model as a first pass editor
- produce a handoff that survives the author leaving
- produce a list of where untrusted input enters a system
- produce a moderation score for an output
- produce a refusal boundary that fires on cases you did not list
- produce a reusable system prompt
- produce a rubric a model can apply
- produce a stopping condition an agent can check on itself
- produce a usable interface for an agent
- produce an account of the gap between ideals and operating values
- produce an inventory of what you pay for and what it holds
- produce an itemised bill for what your running system costs
- recover a working state from your own version history
- segment text into tokens and cost it
- transform a corpus into an embedded index
- transform a transcript into a structured record
- characterise a models failure modes
- characterise a plan's weaknesses with adversarial review
- characterise an applications injection surface
- characterise how a style works on a reader
17 more on this rung
- characterise how an AI assisted attack unfolded
- characterise what a tool costs you beyond its price
- characterise what one user costs you to serve
- characterise where a teams work stalls
- characterise who gains and who absorbs the cost of a deployment
- classify each recurring AI cost by keep downgrade or replace
- classify harmful outputs against a values statement
- classify models by fit for a workload
- classify where AI can fill a shape and where it must not choose one
- classify which framework applies where you are in the lifecycle
- constrain model output to a schema
- generalise a reasoning prompt pattern
- generalise a specification an agent can build from
- generalise a tool interface for model use
- generalise several sources into a position you can defend
- parameterise a prompt template
- parameterise an agent persona
- falsify a claim that a text was written by a person
- falsify a models alignment guardrails
- falsify a prompt with a benchmark
- falsify an applications defences against a named attack class
15 more on this rung
- hold several audience models on one surface
- justify a safety case for a system that acts without you
- justify a vector index for a latency and corpus budget
- justify an embedding model and its dimensionality
- measure a models value alignment against a stated standard
- measure whether a fine tune changed behaviour
- measure whether a user pays more than they cost
- verify a model output you thought was random
- verify a program does what you claimed with a test
- verify a rubric against independent graders
- verify a spending limit declines the charge you did not intend
- verify an agents actions with a critic
- verify model written code against its specification
- verify synthetic data preserves the property you need
- verify you would know your system broke before a user tells you
- construct a multi agent conversation with turn taking
- construct a predictive account of a model's behaviour
- construct a retrieval system
- construct an agent that uses tools
7 more on this rung
- design a chunking strategy against measured retrieval
- design a packaged assistant over your documents
- design a recursive summariser for oversized documents
- design a review gate that catches what the doer cannot see
- design a system you can afford to keep running
- design an evaluation regime that decides model changes
- design an interaction protocol for responsible use
- arbitrate between the measure a system optimises and the goal
- arbitrate which internal feature drives a model's behaviour
- govern a swarm by information hierarchy
- re architect an agent as a stateless reducer
14 more on this rung
- reconcile a failing retrieval by reranking and rewriting
- reconcile a working set with the context window it must fit
- reconcile agents from different toolchains into one run
- reconcile an agent's run with a tool that failed
- reconcile what you built with what you can maintain alone
- select among agent architectures
- select among operating boundaries for inputs a system was not built for
- synthesise a long work that stays consistent across many generations
- synthesise a memory hierarchy for an agent
- synthesise a metalanguage for a problem domain
- synthesise a retrieval over relationships not just similarity
- synthesise a self extending agent behind a review gate
- synthesise an accountability regime for an unattended agent
- synthesise an agent that carries notes across its own runs
- originate an operating model for managed agent teams
- originate oversight that holds when the system outpaces the reviewer
3 of 12 classes fit your 5-point budget (37 hours), and they open 44 of the 99 things above, as far up as write the playbook.
That opens the recordings, the exercises and the written curriculum on this path, and the standups every week. $500 a month adds a seat in every class on it that runs this month.
Or just come to Agentic SDLC, Mon 9 Nov, pay what you can.
What a point is. Ten hours of your time, the recordings you watch, the exercises you work through, and the session itself. Move either slider, or pick the classes yourself and see what the hours come to.
Already on the calendar
The next 2 sessions on this path
The live plan is a seat in each. Bring the traceback you are stuck on.
Two ways in
Watch it all, or be in the room
Same curriculum. The difference is whether you are in the room.
$250a month
Watch it all
- 230.5 hours recorded, yours immediately, and more as each class runs
- Exercises, written curriculum and tools for every class
$500a month
Be in the room
- Everything on the left, plus a seat in every upcoming class on this path, 2 on the calendar
- Bring your own traceback and ask about it out loud
Both renew every month and you cancel either one yourself, any time. A class you complete stays yours after the subscription stops.
Or take the 2 one at a time: $410 total, each pay what you can. Using Large Language Models $60, Agentic SDLC $350.
Own your deployment stack
$750 suggested
Add on anytime, keeps your current path. One-time purchase, classes yours to keep.
The sequence
12 classes, from prompt to deployed system
Each class stands on the one above it.
-
01
Using Large Language Models
Optional. A survey of the field for people new to LLMs, skip it if you already build with them. Class 1 of 12 -
02
Control AI Spending
Class 2 of 12 -
03
AI Alignment
Class 3 of 12 -
04
Context Engineering
Class 4 of 12 -
05
Claude Model Context Protocol
Class 5 of 12 -
06
Intro to Agents
Class 6 of 12 -
07
Prompt Engineering
Class 7 of 12 -
08
RAG & Memory
Class 8 of 12 -
09
Advanced Retrieval Augmented Generation
Class 9 of 12 -
10
Production Agent Engineering
Class 10 of 12 -
11
Agentic SDLC
Class 11 of 12 -
12
Agentic AI Security: Securing What You Build
Class 12 of 12
Straight answers
What is behind the login
The tools that come with the classes
Open them next to your own repo, with the recording paused.
- Using Large Language Models 4 guided workbenches, 3 aids.
- Control AI Spending 1 guided workbench.
- AI Alignment 1 guided workbench.
- Context Engineering 4 guided workbenches, 3 aids. From a bloated prompt to a context budget you can defend.
- Claude Model Context Protocol 1 guided workbench.
- Intro to Agents 2 guided workbenches, 1 aid.
- Prompt Engineering 3 guided workbenches, 1 aid.
- RAG & Memory 2 guided workbenches, 1 companion, 3 aids.
- Production Agent Engineering 1 guided workbench, 1 aid.
- Agentic SDLC 3 guided workbenches, 1 companion, 1 aid, 9 walkthrough decks.
- Agentic AI Security: Securing What You Build 1 walkthrough deck, 1 reference. A map of where untrusted input gets into what you built.
What to expect
What this path asks of you
- You write code and live in a terminal. From Intro to Agents on, you run processes, read stack traces and configure endpoints yourself.
- You end up owning the running system, the bill, the outage and the injection surface.
- One curated sequence, each class standing on the last. See all paths.
Want AI doing your work without building the plumbing? No terminal in these:
One sequence, worked out already
Press play tonight, be in the room on 9 November
230.5 hours open the moment you join, in the order we teach them.
Weekly standups on both plans. Cancel any time.
Two ways in
Recordings $250/month Recordings, exercises, curriculum, standups. Live classes $500/month All of that, plus a live seat in every upcoming class.Full scholarships exist. If the number is the thing stopping you, apply for one.