Your documents and your data.

Privacy notice · Effective October 14, 2026

Sernia Ventures LLC operates Docjig. This notice explains what Docjig processes, who receives it and what happens when you delete it.

What we process

We process your sign-in identity and private workspace; uploaded PDFs and extracted text/images; field labels, types and values; template and document wording; review choices and notes; saved revisions and approvals; editing-copy references; and basic usage and connection records. These documents may contain names, addresses, contact details, financial amounts, signatures and other information you include.

For requested signature collection, we also store sender and recipient contact details, field assignments, consent and signing events, adopted signatures or initials, completed documents and file fingerprints. Private invitation and session credentials control recipient access.

The optional document checker receives only a document record ID and SHA-256 fingerprint calculated in your browser; the PDF is not uploaded. It returns a match result without disclosing document contents or recipients.

Essential cookies and browser storage support sign-in and the interface. Authentication and hosting providers also process technical information needed for their services, such as IP addresses, device/browser details and security events. We receive information you choose to send to support.

Why we use it

We use your information only to provide, support and protect Docjig and meet legal obligations. This includes authenticating you, isolating your workspace, suggesting fields and wording changes, preserving approved versions, creating and exporting documents, delivering requested signing invitations and completed copies, applying usage limits, preventing abuse, providing support and recovering the service.

We do not sell or license your documents or personal information, use them for advertising, or use your content to train AI models. Your workspace is private; we do not publish it as a public document library.

Processing providers

Stripe Managed Payments handles optional subscriptions, payment details, tax, receipts and transaction support. Hosted Checkout collects your payment and billing details directly; Docjig does not receive full card numbers. Docjig stores provider customer, subscription, checkout and event identifiers, payment status and paid-access dates. We send an opaque workspace identifier and the selected plan to Stripe, not document content. Stripe and Link process their records under their own privacy notices.

Clerk handles identity and sign-in. Railway hosts the application. Neon stores private document data and revision history in a US region. Cloudflare provides domain services. Logfire receives operational diagnostics: opaque document/revision/run identifiers, error categories, timings and counts, plus a random receipt for each new account activation. Signup receipts contain no name, email address or sign-in identifier. These diagnostics exclude document text, images, signatures, field values, prompts and model responses.

Resend delivers signing invitations and completion emails to the addresses you specify. It processes email addresses, message content and any attached completed PDF for delivery. Recipients receive their assigned signing access and the completed document and record.

When you request AI field discovery or wording changes on the website, or create or prepare a template through the current MCP connection, the relevant document text, instructions and review evidence are sent through OpenRouter to the configured AI provider, currently OpenAI's Luna model. Native PDF forms with supported existing fields do not need AI field discovery. We use API services configured not to use your content for model training. Providers may temporarily retain submitted content to deliver the service, monitor abuse, protect security or meet legal requirements, under their retention policies. OpenAI explains these limits in its API data controls.

Historical experimental field discovery used TypeSafe's Jev service through OpenRouter. New templates in the current MCP workflow use Luna instead. Existing Jev naming drafts can still be read and their labels saved without a new Jev request. TypeSafe does not train or fine-tune models on those inputs. TypeSafe's privacy policy and data processing addendum describe its service processing and protections.

Authorized service operators may access data when needed to investigate a support or security issue. Providers process information under their own service and privacy terms. We may disclose information when required by law or to protect the service and its users.

Connected AI clients and external copies

When you connect Docjig to an AI client such as ChatGPT, the client can receive the information needed for the tools you use: document titles, selected wording and images, field values, review differences, revision/approval status and document identifiers needed for subsequent actions. The current MCP workflow uses server-side field discovery for new templates; a client can also supply its own field or wording proposals for Docjig to validate and save. That client's own data controls and retention policies apply to its conversations and tool results.

Disconnecting an AI client stops that connection's future access to your Docjig workspace. It does not close your Docjig account or delete your saved documents; use Docjig's deletion controls for that. Downloaded PDFs, exported DOCX files and Google Drive or Word copies are outside Docjig's control. Disconnecting a client or deleting Docjig content does not delete those copies or earlier conversation results.

Retention and deletion

Saved documents remain until you permanently delete them or close the account. Moving an item to Trash keeps it available to restore. Permanent deletion removes live revisions, approvals and editing-copy records. Source and image assets are removed after their last document reference is gone. Templates referenced by saved documents currently require deleting those dependent documents first. Documents with signing records, including cancelled or declined requests, are protected from individual permanent deletion. Account closure removes the account's live document and signing data; it cannot remove copies already delivered to recipients.

Account deletion through Clerk removes the account's live document data after the deletion event is received. We keep an opaque identity/workspace identifier and creation/closure timestamps to prevent old sessions from recreating a deleted account. This closure record remains while the service operates.

Account closure removes document-creation usage receipts and plan enrollment. Minimal payment identifiers, processed-event receipts and cancellation state remain while the service operates so delayed payment events cannot reopen the account or restart access. We do not retain full webhook payloads. Stripe may separately retain transaction and tax records for its legal obligations; deleting Docjig content does not erase those provider records. Contact support for payment-data deletion questions.

Historical database copies can retain deleted information for up to seven days. Inference accounting retains request identifiers, status and timing, without deleted document references; records older than 30 days are pruned on later inference activity. Encrypted connection records have a separate expiry of up to 90 days and are pruned on connection activity. Closed accounts cannot access document tools while those records await pruning.

Operational and support records may remain as needed for troubleshooting, security or legal obligations. Keep your own exports of documents you need; backup recovery is not a document archive service.

Your choices and contact

You can download documents, delete items, close your account and disconnect AI clients. Contact support@docjig.com for access, correction or deletion questions, or to exercise privacy rights applicable where you live. We may need to verify account ownership before acting.

Docjig is intended for adults. Upload only information you have authority to process. We will update this page when our data practices change.