Answers to the obvious questions.

DeployRamp is a feature flag management platform that uses AI to handle the full lifecycle automatically: scan PRs for risky changes, wrap them in flags with tracing, manage the rollout by monitoring errors in real time, and clean up the flag code when rollout completes.

DeployRamp uses AI to analyze your PR diffs for patterns that indicate risk: new API endpoints, database migrations, payment logic changes, authentication modifications, and more. It then suggests wrapping those changes in feature flags so they can be rolled out gradually.

DeployRamp provides official SDKs for TypeScript/JavaScript, Python, Go, Java/Kotlin, Rust, and Swift. The TypeScript SDK includes first-class support for React. DeployRamp works with any CI/CD pipeline.

The DeployRamp SDK reports errors associated with specific feature flags. If error rates spike during a rollout, DeployRamp automatically pauses or reverts the rollout to the safe code path — no redeploy needed. Once the issue is resolved, rollout can resume.

No. DeployRamp automatically creates feature flags from its PR analysis. When it detects risky changes, it suggests a commit that wraps the code in a flag with tracing. You can also create and manage flags manually through the dashboard.

Once a flag reaches 100% rollout and monitoring confirms stability, DeployRamp automatically opens a cleanup PR that removes the flag code, leaving only the new code path. This prevents flag debt from accumulating in your codebase.

DeployRamp supports GitHub, GitLab, and Bitbucket. Cloud-hosted versions of all three are supported on all plans. On-premises (self-hosted) installations are available on the Enterprise plan.

Install the DeployRamp GitHub App on your repository, add the SDK to your project, and you're ready to go. DeployRamp will start analyzing your PRs and suggesting feature flags automatically. The whole setup takes under five minutes.

We don't store source code long-term, no. We will make an ephemeral clone of your repo that is deleted immediately after pushing your commit.

Whenever we are performing an operation on customer source-code it runs in a sandboxed worker that is only running that operation.

Never. We also have a zero-retention and no-training agreement in place with our LLM providers.

Yes. SAML-based SSO is available on the Enterprise plan and above. Contact us if you'd like to upgrade.

We have a trust center that is available to prospective customers of the Enterprise plan.

Yes. DeployRamp is free for teams with up to 3 developers and 1,000 monthly active users. The free tier includes unlimited feature flags, automated flag management, and error-aware rollout.

A developer is anyone who has opened a pull request in a repository where DeployRamp is installed during the last rolling 30 days. Each person is counted once regardless of how many PRs they open or how many repos they contribute to within your organization.

An MAU is any unique context that evaluates a feature flag within the last rolling 30 days. In a browser environment, each distinct browser counts as one MAU. In a backend or server environment, each service instance counts as one MAU; for example, restarting a Kubernetes pod creates a new MAU. Each context is counted at most once regardless of how many flags are evaluated or how often requests are made.

Not immediately. We won't disable flag evaluation solely due to a usage overage or a missed payment without first giving you at least 30 days' written notice. Cutting off flag evaluation mid-rollout could expose your users to untested code paths or break a rollout in progress, which is exactly the kind of disruption DeployRamp is designed to prevent. If you exceed your plan limits or have a payment issue, we'll reach out to resolve it and your service keeps running during that notice period. Accounts that are more than 60 days past due, in material breach of our Terms, or posing a security or stability risk may be suspended without the standard notice period.