Tuesday
Thursday
Wk 1
Tue 08/25
Decide what stays human, then set up the environment agents need
AI makes code abundant; systems judgment and verification become scarce
Authenticate the department shell to the GitHub identity that owns the work
Thu 08/27
How a coding agent acts, and the environment it acts in
Understand the approach before edits; define the evidence before acceptance
Split setup scripts by what they mutate, not by which platform runs them
Wk 2
Tue 09/01
Budget the session's context, then get an independent check on its work
HW DUE dotfiles
Spend the session's working memory deliberately
Make another model challenge the work before you accept it
Thu 09/03
Structure the project's data, then make the agent's v1 teach you the system
Use SQLite when local data must stay structured and durable
A passing v1 is evidence to interrogate, not the finished assignment
Wk 3
Tue 09/08
Contain what a session can reach when you are not watching
The attacker never logs in: you deliver their text to an agent that can act
Decide what the agent can touch and when it must ask
Thu 09/10
Make knowledge outlive the session and every write survive the crash
Loaded memory taxes every future session; retrieved memory costs nothing
Design every write around the crash that interrupts it
Wk 4
Tue 09/15
Pause work without losing it, then prove the database did what you asked
A task is ready to run in parallel when you can name its output and its merge point
Write the query first; the schema and indexes exist to serve it
Thu 09/17
Reason about computation under a hard budget
HW DUE email-index quiz
When exact execution is too expensive, reason about behavior
Allocate a fixed compute budget where it changes the ranking
Wk 5
Tue 09/22
Compare the designs that passed, then set the bar concurrency must meet
Lessons from the email-index homework
Concurrent success must look like a legal sequential history
Thu 09/24
Understand the log that makes a write durable
Topic A to be determined
Two designs buy the same durable instant: log the old page or log the new one
Wk 6
Tue 09/29
Turn a branch proposal into reviewable evidence
HW DUE tm-compute quiz
Make a change easy to understand, test, discuss, and decide
Topic B to be determined
Thu 10/01
Guest lecture by Brent Waters
Guest lecture by Brent Waters
Wk 7
Tue 10/06
Read the rack you budgeted for, then reason about shared execution
Lessons from the tm-compute homework
Correctness depends on every legal thread interleaving
Thu 10/08
Recognize hidden sharing before it becomes the scaling limit
HW DUE tm-compute PR round
Shared state turns memory traffic into the scaling limit
Topic B to be determined
Wk 8
Tue 10/13
Read what review changed, then read the history that preserves it
Lessons from the tm-compute PR round
Immutable snapshots and movable refs explain branches, rebase, merge, and worktrees
Thu 10/15
Agent teams help only when work and interfaces are separable
Fresh context earns an independent judgment only when evidence and role are bounded
Parallel work begins only after the dependencies separate
Wk 9
Tue 10/20
Find the change in a large system, then run code you do not trust
HW DUE concurrent-index quiz
Follow exact names and values until one causal slice explains the symptom
Isolation is a ladder: lower rungs advise, upper rungs enforce
Thu 10/22
Preserve the source and verify the model's claim
Separate source evidence, generated candidates, and derived views by lifecycle
Pretraining predicts text and post-training shapes behavior; neither is understanding
Wk 10
Tue 10/27
Compare the concurrency designs that passed, then expose what the interface never promised
Lessons from the concurrent-index homework
A missing commitment at the boundary becomes behavior every caller must handle
Thu 10/29
Keep credentials encrypted at rest and replaceable when exposed
Encrypted, MACed credentials live in git; plaintext lives only inside one command
Topic B to be determined
Wk 11
Tue 11/03
Change code you do not fully understand, one verified slice at a time
Topic A to be determined
The smallest coherent fix repairs the whole dependency class, not one case
Thu 11/05
Topics to be chosen as the course slate develops
HW DUE pdf2md quiz
Topic A to be determined
Topic B to be determined
Wk 12
Tue 11/10
Learn what the extraction homework cost, then preserve expensive evidence
Lessons from the pdf2md homework
Preserve expensive observations so wrong interpretations stay cheap to fix
Thu 11/12
Decide where automation must stop and ask for approval
An unattended agent gets no permission prompt; approval is code you write
Topic B to be determined
Wk 13
Tue 11/17
Recover when a service call fails, and pay only for results you accept
A stable identity turns retries into convergence on one effect
The cheapest model is expensive when its answer is rejected
Thu 11/19
Topics to be chosen as the course slate develops
Topic A to be determined
Topic B to be determined
Wk 14
Tue 11/24
Thanksgiving break
🦃 Thanksgiving break
Thu 11/26
Thanksgiving break
🦃 Thanksgiving break
Wk 15
Tue 12/01
Demonstrate the system and defend the evidence
DUE Final project
Final project demos 🎤
Thu 12/03
What the projects taught us about working with agents
Final project demos continued 🎤
Slide decks organized by topic
Platform
Accounts
Tools
Working with agents
Homework
Build a search engine for your inboxin progress
tm-compute: protocol & deliverablein progress
Final project
Final project guidein progress