Privacy Policy

This Privacy Policy explains how Release Assurance Inc. ("Release Assurance," "we," "us") collects, uses, and protects information when you use our Salesforce verification platform and this website (together, the "Service"). It aims to be clear rather than exhaustive; if anything is unclear, contact us using the details below.

1. Information we collect

• Account information — the names, company email addresses, and company details of the people you authorize to use the Service. Personal email addresses are not accepted.

• Connection information — the Salesforce sandbox endpoint and the credentials or keys you provide so we can run verification using the account you designate.

• Verification artifacts — screenshots, video recordings, and run logs produced while a test runs. Because tests exercise your real screens, these artifacts may incidentally include data shown on those screens.

• Hosted browser session data — when you use the hosted authoring browser (Section 5), the rendered pixels of your org and the keystrokes and pointer events you send to it pass through our infrastructure while the session is open. They are relayed, not collected: Section 5 says what that means.

• Usage and log data — standard operational and security logs, such as request metadata and error information, generated as you use the Service.

• Website analytics — optional aggregate usage analytics for our website, collected through Google Analytics only after you accept analytics cookies, to understand how the site is used.

2. How we use information

We use information to provide, operate, secure, support, and improve the Service — including configuring and running verification, reporting results, communicating with you about your account, and meeting legal obligations. We do not sell your personal information, and we do not use your Salesforce data to train machine-learning models for others.

3. How we share information

We share information only with service providers who help us operate the Service, and only to the extent needed for that purpose. Our current providers include Google Cloud (hosting and infrastructure), Resend (transactional email), and Sentry (error monitoring). Model requests are processed by Google Vertex AI (section 4). The full list of subprocessors is on our security page. We may also disclose information where required by law or to protect our rights and the safety of others. We do not otherwise share your information.

4. AI processing and model providers

Release Assurance uses large language models to author verification plans from your descriptions, to classify why a run failed, and to answer questions about your testing environment. Where a feature needs it, the content sent to a model can include the description you wrote, the structure of the workflow, and context drawn from a failed run — for example the failing step, captured error text, and a summary of the evidence collected.

Where inference happens. Model inference is performed by Google Cloud Vertex AI. It currently runs in a Google Cloud project administered by our founder in a separate Google Cloud organization, and it moves into Release Assurance's own Google Cloud project before our first customer goes live. We do not send customer content to consumer AI products or free-tier AI services.

Training. Content sent for inference is not used to train foundation models. Google Cloud contractually commits that customer prompts and responses submitted through Vertex AI are not used to train its models, and we do not train models on customer content ourselves.

Minimisation. Captured field values and record identifiers are redacted before a captured step is sent to a model, because the model needs the shape of an action rather than its contents.

Direction. For customers with stricter requirements we are evaluating privately deployed models on dedicated infrastructure, so that inference for those workloads can remain inside a controlled network boundary. That capability is not available today, and this policy will be updated before it is.

5. The hosted authoring browser

Some authoring work happens in a browser that runs in our cloud rather than on your computer. When you start one, we provision a container running Chromium, sign it into the Salesforce sandbox you selected as the principal you designated, and stream what it renders to your screen. Milita drives it; you watch.

What transits, and what does not. Rendered frames and your keystrokes and pointer events are customer data and are treated as such. They are passed through in memory — newest frame only — and are not persisted and not written to any log, including error paths and socket diagnostics. There is no session recording: we keep no copy of what the browser showed you.

Isolation. The container serves one session and is destroyed when that session ends. It holds no long-lived credential, runs its browser as an unprivileged user with Chromium's own sandbox enabled, and can reach the network only through an egress proxy restricted to Salesforce. Downloads are disabled by policy.

Where it runs. In our Google Cloud project, in the United States. Rendered pixels of your org therefore transit our cloud, in memory, for as long as the session is open. If that matters to your data-residency position, see Section 8.

Sandboxes only. As everywhere else in the Service, a hosted browser opens against a Salesforce sandbox or scratch org and is refused against a production org.

6. Data retention

We keep account information for as long as your account is active. Verification artifacts are retained for a limited period and then deleted automatically; the default window is short (7 days) and can be configured for your account. Salesforce access tokens are used only in memory while a test runs and are not stored on disk. When your account ends, we delete or return your data as described in Section 8.

7. Security

We use commercially reasonable technical and organizational measures to protect information, including encryption in transit, access controls, separation of each customer's data, and least-privilege access to the environment you connect. No method of transmission or storage is completely secure, but we work to protect your information and to limit what we collect in the first place.

8. International transfers

We operate on Google Cloud and may process information in the United States and in other countries where our providers operate. Where required, we rely on appropriate safeguards for cross-border transfers of personal information.

9. Your rights and the DPA

You retain all rights to your Salesforce data and business information. Depending on your location, you may have rights to access, correct, export, or delete personal information, and to object to or restrict certain processing. On request we will provide a Data Processing Agreement aligned with the GDPR. When your account ends, we revoke our access and delete or, on request, return your data and artifacts.

10. Children's privacy

The Service is intended for business use and is not directed to children under 16. We do not knowingly collect personal information from children.

11. Changes to this policy

We may update this policy from time to time. When we do, we will revise the "Last updated" date above; we will communicate material changes through the Service or by email.

12. Cookies and analytics choices

The website uses local storage to remember whether you accepted or rejected optional analytics. Google Analytics and its cookies are not loaded unless you choose "Accept analytics." You can change your choice at any time through "Cookie preferences" in the footer. Essential storage needed for security, sign-in, or remembering this choice does not depend on analytics consent.

13. Contact

Questions about this policy, a privacy request, or a Data Processing Agreement: privacy@releaseassurance.com.