Law professor's five-week method turns students from prompters into AI tool builders
jkubicki · x · 2026-09-19
A law professor describes how, in five weeks, students went from one-off chat prompting to building AI tools for themselves.
Key elements:
- Each student builds a tool for their own job (practice-area desk or job hunt), using Jobs Theory to define functional, social, and emotional jobs
- Build an "instrument," not a super prompt: separate instructions, knowledge, behavior, and outputs so each can change independently
- Start with an interviewer that produces a profile file embedding personalization into the instrument
- Define DONE: what the model must learn, a question budget, and what to do with gaps
- Write 3 must-haves and 1 must-never before running; run, check, change one thing, run again—a gateway to evals
More from coding & agent
- LangWatch launches Instant Evals: run any eval across your full production trace history via SQL — _rchaves_ · 2026-09-19
- Xplor, an open-source Grok-native browser built on Chromium, hits 0.8.13 — Daniel_Farinax · 2026-09-19
- Monte Carlo parking simulation with equal cars, built with Claude Code and Marimo — S_Conradi · 2026-09-19
- Two AI Agents Team Up on Their Own: 4,700 Leads Harvested and a Self-Written Contract in 2 Hours — illscience · 2026-09-19
- Reef Adds Tinker API Support: One-Line LoRA Config Brings Continual Learning to Agents — pliang279 · 2026-09-19
- Full prompt revealed: building a DKCR-style 2D platformer with a critic subagent loop — RanaHanocka · 2026-09-19