140 lawyers taught to build their own AI agents in 3 hours, from prompting to personalized analyst briefs
jkubicki · x · 2026-10-25
A 3-hour Chicago workshop walked 120+ lawyers (140 in the room) through building a hyper-personalized 'coverage desk' AI agent. Participants ranged from prompt beginners to deep daily users, but no builders.
Key takeaways: teaching moves from prompt basics to durable, agentic builds; beginners consistently hit a wall treating AI as a magic answer machine rather than something to build with; those who pushed through learned to stitch steps together.
The end product (see image): a 'Monday Desk Brief' tailored to each lawyer's practice — e.g. tracking the 340B rebate model dispute across litigation, legislation (H.R. 10770) and key dates, with Bottom Line / Fresh Signal / Suggested Move sections and working drafts.
Related event: Workshop turns 140 lawyers into AI agent builders in three hours(2 posts)→
More from coding & agent
- 19 latency patterns for faster AI apps: locality, work reduction, concurrency, anticipation — bibryam · 2026-10-25
- 10 open-source repos that turn the web into AI-ready data — Shruti_0810 · 2026-10-25
- popquiz: a free CLI that quizzes you on whether you understand the code AI wrote for you — Honest_Traffic_8613 · 2026-10-25
- Old Android phone + Termux runs 24/7 AI agent, trains PyTorch MNIST to 99.59% — PMinervini · 2026-10-25
- New agent evals field guide draws fire: only 15 of 313 citations predate 2023 — fredahshi · 2026-10-25
- Spotify repositions itself as an AI agent platform with conversational media features — teropa · 2026-10-25