AI and software for law firms
We design and build AI for law firms, notaries and in-house legal teams: search over your own precedents, review assistants that check clauses against your playbook and automation for intake and matter admin. Lawyers review every output, and client files stay in your document management system.
Where the work gets stuck.
Knowledge locked in the DMS
Years of opinions, contracts and memos sit in iManage or NetDocuments, findable only if you remember the client, the matter or the person who wrote them. Associates rebuild work that already exists.
Review work that grows with page count
Due diligence, lease reviews and contract comparisons take hours of careful reading. As more clients ask for fixed fees, those hours become the firm's own cost.
Admin around every new matter
Intake forms, conflict checks, Wwft client due diligence and engagement letters are rekeyed between email, practice management and finance systems before any legal work starts.
How we improve law firms.
Each one is scoped around your systems and rules, and each one keeps a person in charge of the decisions that matter.
Precedent and knowledge search
Lawyers ask a question in plain Dutch or English and get answers from the firm's own opinions, contracts and memos, each linked to the source passage. Search follows the matter security of your DMS, so ethical walls hold.
Contract review against your playbook
The assistant reads an incoming contract, compares each clause with the firm's or client's playbook and marks deviations with suggested fallback wording. The lawyer accepts, edits or rejects every mark before anything goes back to the other side.
Due diligence extraction
Change of control, termination and assignment clauses are pulled from hundreds of data room documents into a review table with page references. Associates verify the table and write the report from structured findings.
Matter intake and conflict preparation
An intake form or email becomes a draft matter in Clio, BaseNet or Legalsense, with parties extracted and run through the conflict search. The responsible lawyer reviews the hits and approves the matter before it opens.
Client portal for documents and status
Clients upload documents, sign through DocuSign and see where their matter stands in one secure place. The secretariat gets fewer calls and emails asking for an update.
Testing legal AI before it goes live
We test legal AI tools against questions your own lawyers have answered, checking that citations exist and quotes match the source. The same test set runs on every change to prompts, models or data.
Built around the rules.
What we design for from the first week. Your legal and compliance people keep the final word.
Professional secrecy and privilege
The duty of confidentiality under the Advocatenwet and notarial secrecy mean client files must stay under the firm's control. We keep documents and indexes in your environment or an EU cloud region, and use model contracts that exclude training on your data and limit retention.
NOvA rules and supplier oversight
The NOvA expects lawyers to stay in control when suppliers process client files, including cloud and AI providers. We document where data goes, which sub-processors are involved and how access is logged, so your firm can assess it.
GDPR and data minimisation
Legal files are full of personal data, sometimes special category data. We redact where possible, limit what is sent to a model and align retention with your file retention policy.
Wwft client due diligence
Where intake automation touches Wwft checks, the software gathers and records the information, and a qualified person completes the assessment and signs it off.
Works with what you run.
If a system has an API, a database, an export or an inbox, we can build on it. These are the ones we meet most.
- iManage
- NetDocuments
- SharePoint
- Clio
- BaseNet
- Legalsense
- Kleos
- DocuSign
- Microsoft Outlook
- Exact Online
Where to start.
Precedent search for one practice group
We start with one practice group, such as corporate or employment, and index its opinions and model documents from the DMS. Lawyers use it on real questions for several weeks while we measure answer quality against their own judgement.
Talk it throughWhat it includes
- Connector to iManage, NetDocuments or SharePoint with permission sync
- Hybrid search with cited answers
- Test set built from questions your lawyers answered
- Usage and feedback reporting
Related work.
Guides.
AI vendor security questionnaire: what to ask before you buy or build
The questions to ask an AI vendor, or your own team, about data, model providers, access, logging, quality, incidents, GDPR and exit, and how to score the answers.
8 min read
AI readiness checklist: what to answer before your first agent goes live
The questions to answer before your first AI agent handles real work, from process and data to approvals, monitoring and ownership.
10 min read
Further reading.
How to build a production RAG chatbot: a practical guide
What it takes to turn a promising RAG experiment into a chatbot people trust: ingestion, hybrid search, citations, permissions and evals.
5 min read
How to evaluate RAG: retrieval metrics, faithfulness and golden sets
How to measure a RAG system properly: separate retrieval from answers, check faithfulness claim by claim and build a golden set you can trust.
4 min read
RAG vs fine-tuning: which does your business actually need?
RAG gives a model your knowledge at answer time; fine-tuning shapes its behaviour. How to choose, when to combine them and what each costs.
4 min read
Common questions.
Documents and search indexes stay in your own environment or an EU cloud region you control. Only the passages needed to answer a question go to the model, under a contract that excludes training on them. Where a matter demands it, we run open-weight models inside your own cloud.
Yes. Both have APIs we use to read documents and their security settings, so the assistant only returns documents the lawyer can already open in the DMS. For files on SharePoint or network drives we build connectors with the same permission check.
They should verify them, and the software makes that quick: every answer cites the document and passage it came from, and citations are checked automatically before an answer is shown. The lawyer remains responsible for the advice.
Most research and drafting tools used by firms fall outside the high-risk categories, which cover AI used by judicial authorities to research and interpret the law. Firms still need staff who understand the AI they use and must tell people when they are interacting with an AI system. We build the logging and documentation that support both.
Often, if the problem is clear. A smaller office on Clio or BaseNet usually gets more from intake and document automation than from a knowledge assistant, because there is less archive to search. We look at where the hours go before recommending a build.