Last updated: January 28, 2026
Plain language summary
Raysurfer stores the code blocks and execution results that your AI agents generate so we can retrieve them later for faster execution. We use third-party services for authentication, payments, search, and analytics. We don't sell your data, and you can delete your account and data at any time.
Account Information: When you sign up, we collect your email address and name through our authentication provider (Clerk). If you subscribe to a paid plan, we collect payment information through Stripe.
Code Blocks: The core of Raysurfer is caching code that your agents generate. We store the source code, descriptions, triggering queries, programming language, dependencies, and metadata you provide when storing code blocks.
Execution Data: When code blocks are executed, we store the inputs, outputs, execution status (success/error), duration, and any error messages. This helps us rank which code blocks work best.
Agent Verdicts: We track "thumbs up" and "thumbs down" feedback on code blocks, including reviews of what worked, what didn't, and suggested improvements. Claude Opus 4.5 may automatically review code blocks for quality scoring.
Usage Data: We track API requests, cache hits/misses, and feature usage through PostHog analytics. This helps us understand how the service is used and improve it.
We use your data to:
Important: We use Claude Opus 4.5 (Anthropic's AI model) to automatically review and score code blocks for quality. This means your stored code blocks and execution results may be processed by Anthropic's API to generate quality verdicts.
This AI processing helps ensure the code blocks returned by our semantic search are high quality and likely to work for your use case.
We share data with the following services to operate Raysurfer:
| Service | Purpose | Data Shared |
|---|---|---|
| Clerk | Authentication | Email, name |
| Stripe | Payments | Payment details, email |
| Pinecone | Vector search | Code embeddings, metadata |
| Fireworks AI | Embeddings | Code descriptions, queries |
| Anthropic | AI code review | Code blocks, execution results |
| PostHog | Analytics | Usage events, feature interactions |
| Resend | Email address | |
| Railway | Hosting | All service data |
Your data is stored on servers in the United States. We use industry-standard security measures including encrypted connections (HTTPS), hashed API keys, and role-based access controls.
Code blocks are stored with namespace isolation—Free tier users share a common namespace, while Pro and Enterprise users have dedicated namespaces to keep their code separate.
Free tier: Code blocks are stored in a shared namespace. This means other Free tier users may retrieve code blocks you've stored if they match a similar query. No personal information is shared—only the code itself.
Pro and Enterprise tiers: Your code blocks are stored in isolated namespaces visible only to your organization. Enterprise users can further segment data by client workspace.
You can:
For data requests, email raymond@raysurfer.com.
We retain your data for as long as your account is active. If you delete your account, we will delete your personal information and code blocks within 30 days, except where we need to retain data for legal or compliance purposes.
We use essential cookies for authentication and session management. Our analytics provider (PostHog) may set cookies to track usage. You can disable non-essential cookies in your browser settings.
We may update this policy from time to time. We'll notify you of significant changes via email or through the service. Your continued use of Raysurfer after changes constitutes acceptance of the updated policy.
If you have questions about this privacy policy or our data practices, contact us at raymond@raysurfer.com.