Before Session 1, name the real system you'll work on for six weeks, and the moment it (or one like it) met production and broke. Everything you build hangs off this.
What this is
Pick the system, and the scar.
This is not busywork. Every deck, every lab, and all five projects are built against your system, so choosing a real one now is the single most important setup step. You return to this reflection in Week 6 and measure the distance.
💡 In plain English
Bring a real problem, not a toy. The course is a six-week workshop on your thing. Someone who brings "the support bot that embarrassed us last quarter" gets ten times more out of it than someone building a generic demo.
Do this, in order
The reflection.
1
Name the system in one sentenceA real or planned LLM feature you own or care about. "A support assistant over our help docs." "An agent that files expenses."
2
Describe the moment it broke (or would)When an LLM feature met production and fell apart: a cost surprise, a confidently wrong answer, an outage, a leak. Be specific.
3
Define "a system that holds"In your words: what would have to be true for you to trust this in front of real users at real volume?
4
State one thing you want to leave able to doYour personal goal for the six weeks. We check it in Week 6.
Fill this in
The one-pager.
My system
one sentence: what it does, for whom
The break
the moment it met production and failed (or would): …
Holds means
I would trust this in production when …
My goal
by Week 6 I want to be able to …
What good looks like
Concrete beats aspirational.
✓ Strong
A real system you can describe and will keep using all course.
A specific failure, with what actually went wrong.
"Holds" defined in terms you could test.
✕ Weak
A generic demo you will abandon by Week 2.
"It was slow sometimes" with no detail.
"I want to learn about AI" with nothing measurable.
Wrapping up
Bring it to Session 1.
Half a page is plenty. This becomes the backbone of every project. Submit on Maven and bring your system to the first live session.
Deliverable
The four-line one-pager. Due Mon Jul 13.
submit on Maven
Not sure yet?
Pick the closest real thing. You can refine the scope in Week 1; you cannot do the course against nothing.
bring something real
For the instructorHow to frame thisclick to open ▾
Set the tone in the welcome email: this is a workshop on their system, not a lecture series. The students who bring a real, slightly-painful system get transformed; the ones who bring a generic demo drift. Nudge waverers toward the most real thing they have, even if it is messy.
Keep it, you will read it back to them in Week 6. The full-circle moment, "remember the break you named in week zero? here is where its answer lives in your design", is the emotional close of the whole course.