Why this memo.
NetDocuments has shipped a credible AI roadmap (PatternBuilder, ndMAX) targeting AmLaw firms with internal innovation teams. Mid-market firms with 20 to 150 attorneys are usually not the right ICP for it. The 47-attorney firm running NetDocuments and a $1.4M annual leakage problem needs custom AI built on top of NetDocuments, not the AmLaw-segment product.
This memo is what we'd commission for that firm. Below: the NetDocuments surface area we touch, the workflows we ship in 5-7 weeks, and how the permissions and ethical-wall architecture stays intact.
The NetDocuments surface area we touch.
NetDocuments exposes the ndMAX REST API for documents, folders, matters, workspaces, profiles, and metadata. Auth is OAuth2 with the firm's identity provider. Permissions and ethical walls are enforced at the NetDocuments server level; the AI layer queries with the user's actual permissions, not a service-account super-user.
This matters more in legal than any other vertical. A retrieval system that bypasses ethical walls is a malpractice incident waiting to happen. Our query architecture preserves them.
Workflow I: Matter-aware retrieval and deal-precedent search.
Year-one associate starts a research task; spends three hours combing the matter archive for prior memos on the same issue, the same judge, the same opposing counsel. Often misses the most relevant precedent because the metadata is inconsistent.
The custom-AI version: associate types the research question in plain English. The retrieval layer queries NetDocuments with the associate's actual permissions, reads matter metadata + document content, returns the top 8-12 most relevant prior matters with the partner who handled each, the outcome, and the specific paragraphs that match.
Recovery: 6-12 months of effective ramp time per associate at audited firms.
Workflow II: Intake-to-DMS pipeline.
New-matter intake usually involves a paper or PDF intake form, a partner triage email, and someone manually creating the matter workspace in NetDocuments with the right metadata. The custom AI receives the intake (form, email, or call summary), runs conflict-clearance against the firm's matter history, drafts the engagement letter from the firm's template, creates the NetDocuments matter workspace with correct profile values, kicks off the document collection from the client.
Recovery: 8-22% of qualified leads that previously went cold during intake delay.
Workflow III: Knowledge ingestion with permissions intact.
The retiring-partner workflow. Twenty years of memos, briefs, deal precedents in the partner's NetDocuments workspaces. We bulk-ingest into a permissions-preserving retrieval index, with the original ACLs maintained. The next lateral hire has access to the firm's intellectual capital, gated by the same ethical-wall logic that gated the original documents.
What we don't build.
We do not replace NetDocuments. We do not bypass ethical walls. We do not build a competitor to ndMAX or PatternBuilder. We do not build "Harvey for your firm" because Harvey already exists and serves a slightly different segment well. The leverage is in firm-specific workflow integration, not generic legal AI.
Risk & governance.
Three things every legal AI commission must do, by default: (1) preserve client confidentiality through actual permissions enforcement, not policy memos; (2) provide an audit trail that survives subpoena; (3) avoid hallucination on dispositive citations through retrieval-grounded generation. Architecture follows from these constraints.