
Discovery Response Drafterflagship
A "selection, not generation" demo for drafting California discovery-response objections — with a chain inspector that shows every move.

Senior Legal Operations & AI Advisor, LECG · Attorney & Legal Engineer
Designing and deploying governed AI workflows for in-house legal teams
Request an introI'm a California attorney and legal engineer. In September 2026 I joined LegalEng Consulting Group (LECG) as Senior Legal Operations & AI Advisor. I work with in-house legal teams to find the AI use cases worth building, design the workflows, and get them deployed and governed inside the tech stack the team already runs. I've practiced from every seat at the table: BigLaw associate at Goodwin Procter, in-house counsel at Cochlear, founder of a privacy and AI boutique, and litigation defense at a firm where I built the AI function from scratch. The first real system was a discovery-response drafter on a selection-not-generation design: an attorney-curated objection library, deterministic rendering, and a verification pass, so the model selects vetted language instead of inventing it. Its output was attorney-reviewed and served on live matters. From there I worked through the rest of the litigation lifecycle on the same propose, review, commit pattern: pleadings, motion practice, deposition prep, and exposure analysis. The layer that decides whether a legal AI tool ever gets turned on is governance: privilege handling, conflicts checks, tamper-evident audit trails, prompt-injection defense, and fail-closed de-identification. Drafting is the easy part; the work is output trustworthy enough to put in front of a partner and defensible enough to clear a GC's security review. I hold both tiers of Anthropic's Claude Certified Architect certification (Foundations and Professional), the IAPP AIGP and CIPP/US, and I'm a LegalQuants resident. I'm most interested in talking with people shipping real legal tooling on either side of the in-house/firm line, and with legal ops teams deciding what to buy, what to build, and what to leave alone.

A "selection, not generation" demo for drafting California discovery-response objections — with a chain inspector that shows every move.
A Claude-built pipeline that ingests a client's payroll and timesheet PDFs in a class/PAGA wage-and-hour case and computes non-compliance exposure for the whole class — work traditionally done by a paralegal typing a subset into a spreadsheet.
© 2026 LegalQuants · legalquants.com · the original home of legal quants.