Legal AI from Pattern Automation v0.1.0: the first public release
The first public version: a self-hostable legal AI platform for document review, drafting, research, and reusable workflows.
Discuss this post in AI
Send a pre-filled prompt to ChatGPT, Claude, Gemini, or Perplexity — get a summary, ask follow-ups, or compare ideas from this guide.
Legal AI from Pattern Automation v0.1.0 launched as a working legal document assistant rather than a demo or a promise. From its first release, legal teams could upload documents, work with an AI assistant, organize matters into projects, run structured reviews, and reuse legal workflows.
The release also made the application runnable for firms that want to inspect operation, self-host, and adapt it to their own legal work — documents on their perimeter, not a paste into a vendor prompt box.
The core features in v0.1.0
The first release brought the essential parts of Legal AI from Pattern Automation together in one application. Users could move from a single-document question to a project-wide workflow or a multi-document review without leaving the platform.
- Assistant — chat with uploaded legal documents, attach source material to a conversation, and receive answers grounded with citations
- Projects — create matter-specific workspaces that bring documents, assistant conversations, and reviews together
- Tabular review — extract structured answers across multiple documents, add custom review columns, trace answers to citations, and export results to Excel
- Workflows — create, edit, save, and share reusable legal prompts and review templates
- Document handling — upload PDFs and office documents, convert DOC and DOCX files for review, and manage files in object storage you control
- Model choice — run the assistant with supported Anthropic, Gemini, OpenAI, or OpenRouter-backed models
Built to be owned and extended
v0.1.0 included the application, API, database schema, authentication flow, and object-storage integration needed to run the product. Teams could deploy it against their own database and storage rather than sending legal documents into a closed vendor environment.
That foundation set the direction for the releases that followed: stronger document workflows, broader model support, safer deployment defaults, legal research, connectors, and a more refined interface.
Harvey is valued in the tens of billions. Legora raised at a multi-billion valuation. The market proved demand. Legal AI from Pattern Automation is the legal function on Neuro OS — the pack, the citation, the workflow — on machines the firm controls.