Turn your
law firm AI-native
Referent is an AI-native practice management software for law firms. Legal AI agents handle the operational work - you approve key actions and focus on clients, relationships, and growth.
Referent is an AI-native practice management software for law firms. Legal AI agents handle the operational work - you approve key actions and focus on clients, relationships, and growth.
Team backed by:
Google for Startups
Nvidia Inception Program
Ready when you are.
Forget legacy legal software where every client, matter, and task has to be entered by hand. Just tell Referent what needs to happen - our Legal AI handles the work.
Referent runs the pre-matter workflow for you. It collects context, organizes details, prepares the proposal, and sends the payment link.
Connect Gmail, Google Calendar, and Google Drive: Referent reads your correspondence and sends the replies you write so every email lands in the right case file, syncs hearings and meetings both ways, and makes your Drive documents searchable inside your matters. Disconnect anytime — synced data is deleted.
Legal AI agents update matter records, create tasks, route work, and keep your legal CRM organized automatically.
Referent links emails, files, notes, and calls context to the correct matter automatically.
Referent does the work, but you stay in control. Every important
action is verified by you before it
happens.
Full firm leverage
Operate like a modern law firm - even if you're a team of one. Let AI agents handle routine work while you focus on practicing law.
Same team. More matters
Turn your firm AI-native. Equip every lawyer with agents that standardize workflows and automate operations.
Scale operations
Roll out an AI operations layer across practice groups. Every action is logged, auditable, and approved by your lawyers.
Workspace setup · Step 4 of 5
Review the results and confirm import to create them in your workspace.
Referent moves every client, matter, task, and file from your existing tools, so your practice is ready on day one.
Works with
Client confidentiality is the foundation of legal practice. Referent keeps your firm's information private, isolated, and under your control — so you can use AI without giving up ownership of your data.
Our vision
Lawyers bring judgment, strategy, and care. AI brings speed, scale, and consistency. Referent exists to put the two together - so a solo or small firm can carry far more matters without losing the craft that makes the work worth paying for.
I spent fifteen years building technology, time inside a law office, and four years in legal tech. The pattern never changed: capable lawyers buried in operational work that software should have handled. Referent is the system I wished existed - it runs the routine and hands every client-facing decision back to the lawyer.
George Zaletski
Founder & CEO, Referent
An AI-native law firm is a firm where AI is built into daily operations, not just opened occasionally to draft a letter or summarize a document. In an AI-native firm, client emails become tasks, new intake turns into matters automatically, documents become summaries, and deadlines create their own reminders.
AI agents move the routine work forward in the background, while the lawyer reviews and approves every action that touches a client or carries real risk. The distinction is simple but decisive: an AI-assisted firm uses AI as a tool it picks up and puts down; an AI-native firm runs on AI as its operating layer, the way a modern business runs on its accounting system.
The practical result is leverage - the same lawyer and the same staff can carry far more matters without losing track, because the operational overhead that used to consume the day now runs itself.
Becoming AI-native takes three steps, in order. First, connect your firm's context - bring clients, matters, emails, documents, and deadlines into one system, so the AI works from your real practice rather than a blank prompt.
Second, hand your repetitive operations to AI agents: intake, follow-ups, task creation, deadline tracking, document filing, billing prep, and client updates are the usual starting points. Third, keep approval in the lawyer's hands - set rules so nothing client-facing or high-risk happens without your review.
Most firms begin by automating those seven routine workflows, prove the system on them, then widen the scope as trust builds. You do not rebuild your practice overnight, and you do not hand over judgment. With Referent, white-glove onboarding handles the setup and the connection work for you, so a firm can reach a working AI-native baseline in days rather than the months a self-serve tool would take.
Referent is built specifically for the transition to AI-native operations, not as a chatbot bolted onto legacy practice management software. It combines two layers that usually live in separate products.
The first is a legal CRM that holds your clients, matters, tasks, documents, and billing in one place. The second is an AI execution layer - voice commands and agents that actually move routine work forward, rather than just answering questions about it. Every agent works from your firm's live matter context, so it acts on your real clients and deadlines, and every critical action waits for a lawyer's approval before it leaves the firm.
Most legal software is a system of record: it stores what already happened and waits for you to do the next thing. Referent is a system of action - it makes the next thing happen, then asks you to approve it. That difference is why firms going AI-native choose it.
Referent's legal AI agents work from your firm's matter context - the clients, emails, documents, tasks, deadlines, and billing status already in your workspace. Each agent owns a specific routine: one handles intake, another routes and files client email, another drafts follow-ups, another tracks deadlines, another prepares billing.
The agent does the legwork and then stages the result for your review rather than acting on its own authority. You can trust them because of how they are constrained, not because you are asked to take them on faith. The rule is constant: AI prepares, the lawyer approves.
Any client-facing or high-risk action requires your explicit sign-off, every action is recorded in an audit trail you can inspect, and every draft can be edited or rejected before a word leaves your firm. The agents remove busywork and create leverage, but the professional judgment - and the final decision - stays with the lawyer.
Referent connects only to the Google services you choose, so your firm's operations can run in one place instead of being scattered across tabs. With Gmail access, Referent files incoming client correspondence to the right matter and sends the replies you have approved.
With Google Calendar access, it syncs hearings, meetings, and deadlines in both directions, so nothing lives in only one calendar. With Google Drive access, it makes the documents you connect searchable inside the related matter, where you actually need them.
The access is scoped to what each feature requires, and you grant each service deliberately. Referent uses Google data only to provide these features - never to train AI models, and never for advertising - and you can disconnect any service at any time, at which point the data synced from it is deleted. Confidentiality is the constraint the whole integration is designed around, not an afterthought.
Your firm's data lives in a secure, encrypted workspace that is isolated per firm, so one firm's information is never mixed with another's. Referent does not train AI models on your client data, and it does not sell, share, or monetize client information in any form.
No client-facing action happens without a lawyer's permission, and you control internal access through roles and permissions, so each team member sees only what their role allows. A full audit trail records what was created, changed, or sent, and by whom, which gives you a defensible record for your own compliance and ethics obligations. When you delete data, it is genuinely deleted, not archived out of sight.
Confidentiality is not treated as an add-on feature here - it is the design constraint everything else in the product is built around, because for a law firm, client trust is the asset that matters most.
A blank AI chat knows nothing about your firm - you paste context in, copy the answer out, and update your own systems by hand afterward. Referent works from your firm's live matter context instead: the clients, emails, documents, deadlines, tasks, and billing already in your workspace.
Ask it 'What needs my attention today?' and it answers from your actual practice, not from generic training data - and then it does the work rather than just describing it. It creates the matter, drafts the follow-up for your approval, files the document to the right place, and sets the reminder.
A chatbot is a smart assistant you have to direct turn by turn, and that re-enters everything manually. Referent is an operating layer that already holds your firm's state and moves routine work forward on its own, while keeping every client-facing action behind a lawyer's sign-off. Chatbots answer questions. Referent runs the operation.
Private beta for selected lawyers and law firms.
Personal invite
White-glove onboarding
No data training