Uber Built a Legal Redlining Agent for Its Lawyers—Here's What It Learned
AxSaucedo · x · 2026-10-28
Uber's engineering team published a detailed postmortem of its Legal Redlining Agent (LRA), started in early 2024 to help legal teams handle thousands of annual contract negotiations with heavy repetitive redline work. The project contributed to Uber Legal winning ALM Legalweek's Most Innovative Legal Department of the Year award.
Key lessons:
- Start from the business problem: target high-volume, repeatable redlining, not general-purpose legal AI;
- Meet lawyers where they work: flag risky redlines and suggest edits inside Microsoft Word, no workflow change required;
- Iterate closely with lawyers: piloted with the legal team in 2025, folding feedback in—earning trust mattered more than replacing lawyer judgment;
- The post closes with the first-gen architecture and how they'd rebuild it today given modern agent harnesses.
The most durable lesson wasn't about the model or architecture, but bringing an agent into lawyers' day-to-day work.
More from coding & agent
- Parallel launches Decisions API: web-grounded yes/no answers with confidence scores at $0.001/request — heyshrutimishra · 2026-10-28
- Investor Gavin Baker: Grok Bot is the most impactful AI product at his firm, with 17 bots built — elonmusk · 2026-10-28
- Wes Roth's Blood Bench: AI builds and coaches competing fantasy football teams end to end — Wes Roth · 2026-10-28
- Claude reads my social feeds nightly and shows highlights on an Apple TV dashboard — florianvo · 2026-10-28
- Developer slams Apple: each iOS simulator stores its own Siri model, wasting 25.5 GiB — Dimillian · 2026-10-28
- Show HN: RSI-Gear, a Lego-like composable framework for recursive self-improvement — ethan1998 · 2026-10-28