Projects

Matteria.

From client intake to ready legal documents.

Matteria is an AI-powered matter intake and document automation platform for law firms. Clients share their information once through a secure portal. Matteria extracts the key details from their documents, autofills the firm’s own templates, and keeps everyone updated in one place, with an attorney approving every document before it leaves.

Matteria dashboard with the matters list, client and legal status, and the messages panel.
Client
Client project
Industry
Legal technology. Launch focus: high-volume personal injury firms.
Services
Custom Software & Product Development
AI & Automation
Enterprise AI & Ongoing Support
Deliverables
SaaS web application, with a staff workspace and a client portal
AI document extraction and template autofill
Billing and onboarding
Stack
  • Next.js
  • React
  • TypeScript
  • Tailwind CSS
  • MongoDB
  • AES-256-GCM
  • Auth0
  • Stripe
  • Google Gemini
  • pdf-lib
  • PostHog
  • Postmark
  • DigitalOcean
  • Cloudflare
  • Vercel
  • Vitest
  • Playwright
  • GitHub Actions
Status
Live
Links
trymatteria.com

The starting point.

  • Clients repeat themselves. The same facts show up across calls, email, intake forms, retainer packets, insurance forms, and authorizations.
  • Staff re-type them. That data is entered again into every document and system the matter needs. The work is repetitive, easy to get wrong, and not billable, and every matter requires it.
  • Updates turn into phone calls. When clients cannot see what is happening, the gaps fill with calls and follow-up email.

The brief was to close the gap between “the client signed up” and “the file is open, with the paperwork complete.”

What we built.

One secure link to start the matter

Staff open a matter and share a single intake link. The client creates an account, follows a guided intake, and uploads documents once, including by camera on a phone. A live checklist shows what has arrived and what is still missing. Requests and reminders go out on their own.

Three Matteria panels: client onboarding, a retainer agreement being autofilled, and a message thread.
Onboarding, autofill, and the shared thread.

AI that turns uploads into structured data

  • Extract. Google Gemini reads uploaded PDFs and pulls out names, dates, addresses, insurers, claim numbers, and related fields.
  • Review. The results land on a single matter profile, where staff can check and edit each field next to the document it came from.

Autofill the firm’s own templates, with a person in the loop

  • Fill. Firms upload their own fillable templates. Matteria maps the approved matter data onto each template and fills it in a couple of clicks.
  • Approve. Filled documents stay pending until an attorney approves them. Staff review highlighted fields and compare versions before anything is exported.

One record and one conversation per matter

A timeline logs every event from the moment the matter opens to the latest upload. A shared message thread keeps client communication on the matter, instead of across inboxes.

Matteria’s six-step intake, from creating an account and a document checklist through extraction, a matter profile, template autofill, and staff approval.
The intake path, in six steps.
Diagram from the Matteria site: client emails, intake forms, insurer details, and other inputs converge in Matteria, then continue to a completed file, autofilled documents, and a notified client.
How scattered inputs become a client file.

Security built in.

Legal files are confidential, so security was part of the foundation. Every document, extracted field, and template is envelope-encrypted with AES-256-GCM and a key per matter. The application refuses to run without its master keys.

Access is role-based: firm admin, attorney, paralegal, intake staff, and client, scoped to each matter. Sessions are hardened, requests are rate-limited, and an audit trail records what happened. Every AI request is logged.

Ready for real customers.

Firms can sign up with Auth0, start a trial, and subscribe through Stripe. Transactional email and product analytics, with masked session replay, sit alongside the app. It is deployed on DigitalOcean behind Cloudflare. The release pipeline runs lint, type checks, unit and end-to-end tests, security scans, and CodeQL.

Next UniWizard. Read the case study

Let's talk about the work on your team's plate. Tell us what you'd like to build, fix, or automate. We'd like to hear about it.

Tell us about your project