Lawyer builds agentic interview tool that maps legal workflows for AI pilots
jkubicki · x · 2026-10-18
jkubicki, a legal-tech practitioner, argues most law firms buy AI before understanding how individual partners actually practice. He built a workflow discovery tool: an agentic interviewer walks a lawyer through their work, helps pick one workflow to pilot or improve with AI, and looks past efficiency toward new value ("alpha").
Output is a visual workflow map plus a detailed breakdown showing where AI fits. The sample image maps the "refreshing sponsor-specific memory before a new deal" workflow—from receiving LOI/CIM through replaying past deals, kickoff briefings, and mid-negotiation lookups—with 6 numbered signposts marking AI opportunities.
The tool takes 20+ minutes of thoughtful answers ("garbage in, garbage out") and was just shared with his newsletter readers; he'll DM links on request.
More from coding & agent
- Engineer watches Opus55 optimize a Jai rewrite: "100x slower by hand" — lucasmeijer · 2026-10-18
- px0 browser IDE adds inline file editing, commit browsing and diffs — arpit_bhayani · 2026-10-18
- Netflix engineer open-sources Headroom, cutting LLM token usage by up to 95% — VeryWellVersed · 2026-10-18
- hubspot-auditor: open-source read-only MCP server runs 46 checks on a HubSpot portal — CryptBay · 2026-10-18
- Map failure modes on paper before automating any workflow, six lines at a time — usehive · 2026-10-18
- Lessons building a session watcher for 7 AI coding agents: hooks beat process detection — Weak_Cherry_4878 · 2026-10-18