catch the citation ai invented
the lawyer path teaches you to build document-review and research tools, and to verify ai output rigorously, because a hallucinated citation is a malpractice risk, not a bug.
who this is for
you practice law or run a legal team. ai already drafts, summarizes, and cites. you need to build tools around it and, more than anyone, verify what it produced. a confident wrong answer in this work is not a small thing. this path gets you there.
what you'll be able to do
build document-review and research tools that work over your actual documents, not the model's memory. get structured summaries out of long files. verify ai output rigorously and catch invented citations, names, and dates. govern how ai is used across a practice, and handle sensitive data the right way.
the route, in order
start in legal studio by building a document-review workflow with citation checks, privilege boundaries, and an audit trail. the shared builder spine supplies the retrieval, structured-output, and evaluation skills that make the review defensible. self-paced; the full route is listed below.
- 01legal studio
- 02from worker to builder
- 03read the script
- 04read the data shapes
- 05read the failure
- 06read the pipeline
- 07llm apis
- 08structured output
- 09prompting cursor and claude code effectively
- 10context and retrieval
- 11eval-driven ai development
- 12debugging broken ai output
- 13secrets
- 14claude skills for teams
how it works
no videos, no fizzbuzz. every step shows a small piece of ai-written code. you read it, predict what it does, find the bug, fix it, run it. it all runs in your browser.
where you end up
builds doc-review and research tools, verifies ai output rigorously, governs ai access for a practice.
free to start, no signup. plus unlocks every chapter at once; the free path moves through them in order. teams get progress visibility and role-fit rollout — see for teams.