
AI lessons, ready for your class.
Choose readings, labs, quizzes, and agent projects from the library and assign what fits. Then open a submission to read the student's answers and follow every step their agent took.
Start with one class. Add the next chapter when you are ready.
Language models
3Build a research agent
18 of 26 students complete · 3 waiting for review
Maya R. · Submitted agent
Why did you add a second tool?
My first run made up a source, so I gave it search and told it to quote the page it found.
model turnResearcher → Model
Asked for the Search tool
tool callResearcher → Search
search("tide pool species California") · 4 results
tool callResearcher → Search
Timed out after 8s. Maya changed the query and reran.
message sentReply
Quoted two sources.
Reviewed records that you looked. It is not a grade.
Choose what fits your next lesson.
Readings, quizzes, interactive labs, and agent projects are already written. Assign one piece or a whole chapter, and it shows up for your class.
Readings
Quizzes
Labs
Agent projects
Students get the published version you assigned, so a later edit never rewrites work already underway.
Classroom AI credits are included in your plan.
Students build on the same models a professional uses, through a visual builder rather than code. Your plan includes a monthly credit balance for those runs. You hold it for the class, so no student signs up for a model account and nothing is billed to you per run. Credits refill each month and do not roll over.
See what each plan includesClass balance
AP Computer Science · 26 students
Credits left this month
24,180of 30,000
Recent runs
- −512
Build a research agent
24 students, 61 runs
- −143
Temperature and sampling lab
26 students
- −64
Tokenization warm-up
26 students
Students pay nothing
You are not billed per run
Unused credits do not roll over
Know who needs a look before the due date.
The progress list puts every student against the chapter, with a plain reason next to anyone who has stalled: no activity, failed runs, a lab never opened. Open their work from there.
Progress
Language models chapter · 26 students
Priya S.
12/12On track
Maya R.
9/12On track
Leo M.
6/123 failed runs on the project
Devon K.
4/12No activity in 6 days
Sana A.
2/12Has not opened the lab
A flag is a reason to open the work, not a grade.
Privacy enforced on every model call.
Trace rewrites the request and sets retention on it before it reaches a model provider. Here is the same message on both sides of that line.
What a student types
Inside Trace
Hi, I'm Maya Chen. Can you check the notes I emailed from [email protected]? My student ID is 4471029.
What reaches the model
After Trace rewrites it
Hi, I'm [REDACTED_STUDENT_IDENTIFIER]. Can you check the notes I emailed from [REDACTED_EMAIL]? My [REDACTED_STUDENT_IDENTIFIER].
Sent on the same request
"provider": { "data_collection": "deny", "zdr": true }The provider is told not to keep the request or use it for training.
Read the exact wording in our Privacy Policy.
Your first class in three steps.
Pick a first chapter for the students and subject you already teach. Start small and adjust from there.
- 1
Create a class
- 2
Assign a first lesson
- 3
Open the work
Common questions
What teachers ask before they set up a class.
No. You assign published readings, labs, quizzes, and projects, then guide the discussion. Students build agents by connecting blocks, so nobody writes code unless a lesson asks for it.
Start with the class in front of you.
Trace gives you the material to assign and a record of what each student did.
Educator starter is free: 2 classes, 30 students each, and 2,000 credits a month.