Senior Counsel
Automating the 80/20 principle
I'm Senior Counsel at Clipboard Health, where I wear multiple hats as head of Commercial, Product, and Compliance. Before this I spent 8 years as a federal prosecutor, and earlier in my career worked in-house at Facebook and as a Privacy and Cybersecurity associate at Holland & Knight LLP. . I believe lawyer-built tooling is the next frontier of the in-house profession. For years, customized legal tech meant a vendor contract, an engineering ticket, or a long wait for something that is tailored to how your business operates. AI has collapsed that loop. A lawyer who understands the business, the risk, and the workflow can now build the tool themselves, faster, and specific to the use case.
The Problem: Legal risk at tech companies is usually delivered in narrative form, "this is risky, consider mitigations," which leaves business teams without the quantified inputs they need to weigh legal exposure against product value. Product managers and executives asking "should we launch this feature?" get qualitative hedging where they need dollar ranges, probabilities, and comparables. Meanwhile, the research required to price a risk accurately (settlement history, AG and federal enforcement patterns, pending legislation, peer company outcomes, political climate) is scattered across dozens of repositories and typically compressed into one lawyer's gut read. The Tool: The Legal Risk Pricer is a multi-agent orchestrator that turns a scenario (a feature, policy, contract position, or regulatory change) into a quantified risk assessment non-lawyers can act on. It runs a three-layer intake to build a confirmed Research Brief, then deploys six parallel research subagents against mandatory repositories (settlements, state AGs, federal enforcers, state legislation, industry news, and comparable companies), consolidates their findings into a dossier, and feeds it to a dedicated scoring agent that produces structured JSON across four dimensions (vulnerability, damage magnitude, enforcement likelihood, asymmetry) and three complementary formulas (expected value, structural exposure, severity × likelihood). A mandatory QA agent audits for arithmetic errors, hallucinated citations, and framing before the orchestrator assembles a Slack-ready summary, an interactive HTML report with mitigation modeling and enforcer cascades, and a persistent risk register for portfolio-level comparison, giving business teams the same kind of pricing rigor finance applies to capital allocation.
An AI-powered operating system for in-house legal teams. Centralizes escalations, playbooks, and knowledge, and gets smarter every week from the work the team does. How it works Slack /escalate command kicks off any matter — no new tool to learn for requesters AI triages every matter: classifies practice area, sets priority, routes to the right attorney Pre-analysis runs before a lawyer touches the matter — pulls relevant playbooks, knowledge base articles, prior similar matters, and counterparty memory into a synthesized brief Drafting workspace composed from a versioned clause library; export to Notion or Google Drive Matter-scoped Copilot has the full context (playbooks, KB, prior matters) and can be invoked anywhere The self-learning loop: Seven weekly cron jobs mine the team's actual decisions to update the institutional brain: Rejection mining clusters lawyer disagreements and proposes new playbooks or routing patches Playbook-edit mining compares matched playbooks against final lawyer-shipped output and proposes targeted amendments Conflict detection scans for duplicate canonical clauses, rule priority collisions, and near-duplicate playbooks Every accepted lawyer edit becomes the next version — the corpus compounds Architecture: Stack: Next.js · tRPC · Drizzle ORM · PostgreSQL · Railway Voyage-Law-2 embeddings across knowledge, playbooks, templates, rules, and execution patterns — global ⌘K semantic search Natural-language rule engine (SLA · routing · triage · playbook triggers) with a four-state lifecycle: draft → shadow → active → archived Codified statutory-research workflow with citation verification — designed to eliminate the hallucinations of pure-LLM legal research Integrations Slack (trigger), Salesforce (matter context), Notion (canonical store), Google Drive, GitHub, Anthropic API (Copilot · analysis) Status: proof of concept, deployed and operational.