Referent is an AI-native legal practice management software, positioned as a law firm's "system of action" rather than a traditional system of record. It is aimed at law firms that want to integrate AI into their daily operations, especially solo attorneys and small boutique firms.
Purpose
Referent is designed to enable law firms to operate AI-natively: AI agents advance daily affairs in the background while attorneys focus on clients, relationships, and business growth. It solves the problem of attorneys being overwhelmed by operational work and traditional legal software requiring manual entry of every record.
Key Capabilities
- Chat-based operations: Users tell Referent what needs to be done via chat, the agent executes, and the attorney approves.
- Client intake: Clients converse with AI through an intake link, and Referent screens for fit, opens the matter, and prepares engagement materials for attorney approval.
- Email filing: After connecting an inbox, Referent files emails, attachments, and context into the correct matters and flags items that need attention.
- Document drafting: Replies, letters, and documents are pre-drafted based on the full context of the matter, and the attorney edits, approves, and sends.
- Deadline tracking: Monitors court dates, filing deadlines, statutes of limitations, etc., and alerts before they become urgent.
- Billing preparation: Records time during work, automatically generates hourly or flat-fee invoices, and attorneys review bills rather than timesheets.
- Follow-up reminders: Payment reminders, client check-ins, review requests, etc. Referent drafts each outreach, and nothing is sent without approval.
- Approvals and audit: Any client-facing or high-risk action requires explicit attorney approval, and all actions are recorded in an auditable log.
- Mobile: Offers iPhone and Android apps to review agent work, approve key actions, and view matters at any time.
- MCP server: Connects to compatible apps such as ChatGPT, Claude, Perplexity, Hermes, Gemini, and OpenClaw through the Model Context Protocol, handling clients, matters, and documents in chat, with permissions, approvals, and audit trails kept consistent.
- Data migration: Migrates clients, matters, tasks, and files from existing tools so the firm can be up and running on day one.
How to Use
Referent is currently in private beta for boutique law firms, requiring an application for access and personal onboarding guidance. The usage process includes: connecting firm context (clients, matters, emails, documents, deadlines), handing off repetitive operations to AI agents, and retaining attorney approval authority. Supports connecting Gmail, Google Calendar, and Google Drive through a Google account, with permissions granted by feature scope, and you can disconnect at any time and delete synced data.
Use Cases
- Solo attorneys: Operate like a modern law firm, with AI handling routine work.
- Boutique law firms: Give every attorney an agent, standardize workflows, and automate operations.
- Multi-practice-group law firms: Deploy an AI operations layer where all actions are recorded, auditable, and approved by attorneys.
Limitations and Data Security
- Currently in private beta, requiring an application for access.
- Does not train AI models on firm data, and does not sell or monetize client information.
- Encrypts and isolates workspaces by firm, with data at rest encrypted using AES-256 and data in transit using TLS 1.2+.
- SOC 2 and ISO 27001 certifications are in preparation.
- Key actions require attorney approval, and all AI actions are auditable.
- Role-based access control, and deletion means deletion.
- Website analytics only measure website visits and do not observe or analyze work inside Referent.
Team Background
The Referent team is supported by Google for Startups, the Nvidia Inception Program, and OpenAI Select Partner. The founder and CEO is George Zaletski.repetitive operations to AI agents, and retaining attorney approval authority. Supports connecting Gmail, Google Calendar, and Google Drive through a Google account, with permissions granted by feature scope, and you can disconnect at any time and delete synced data.
Use Cases
- Solo attorneys: Operate like a modern law firm, with AI handling routine work.
- Boutique law firms: Give every attorney an agent, standardize workflows, and automate operations.
- Multi-practice-group law firms: Deploy an AI operations layer where all actions are recorded, auditable, and approved by attorneys.
Limitations and Data Security
- Currently in private beta, requiring an application for access.
- Does not train AI models on firm data, and does not sell or monetize client information.
- Encrypts and isolates workspaces by firm, with data at rest encrypted using AES-256 and data in transit using TLS 1.2+.
- SOC 2 and ISO 27001 certifications are in preparation.
- Key actions require attorney approval, and all AI actions are auditable.
- Role-based access control, and deletion means deletion.
- Website analytics only measure website visits and do not observe or analyze work inside Referent.
Team Background
The Referent team is supported by Google for Startups, the Nvidia Inception Program, and OpenAI Select Partner. The founder and CEO is George Zaletski.
