Subprocessors
On this page
Last updated: August 17, 2026
Wonder (operated by Aquila Labs, Inc.) engages a limited set of third-party subprocessors to help us deliver our service. This page lists all subprocessors that may process customer personal data, along with the purpose of processing and the location where data is hosted.
We will notify customers of any changes to this list by updating this page. Customers can subscribe to updates by emailing team@wonder.so.
Infrastructure and hosting
| Subprocessor | Purpose | Location | Website |
|---|---|---|---|
| Vercel Inc. | Website hosting, Web Analytics, and Speed Insights | United States | vercel.com |
| Google Cloud Platform (Google LLC) | Primary cloud infrastructure, compute, networking | United States | cloud.google.com |
| Firebase (Google LLC) | Application backend services, real-time data sync | United States | firebase.google.com |
| PostgreSQL (managed via GCP) | Primary relational database | United States | postgresql.org |
| Redis (managed via GCP) | Caching and session storage | United States | redis.io |
| Prometheus | System metrics collection (self-hosted on GCP) | United States | prometheus.io |
| Grafana Labs, Inc. | System metrics visualization and alerting | United States | grafana.com |
| ToDesktop | Desktop application build, distribution, and auto-update | United States | todesktop.com |
Authentication and payments
| Subprocessor | Purpose | Location | Website |
|---|---|---|---|
| WorkOS, Inc. | Authentication, SSO, and user directory | United States | workos.com |
| Stripe, Inc. | Payment processing and billing | United States | stripe.com |
AI model providers
| Subprocessor | Purpose | Location | Website |
|---|---|---|---|
| Anthropic, PBC | Large language model inference for AI-assisted design features | United States | anthropic.com |
| OpenAI, L.L.C. | Large language model inference for AI-assisted design features | United States | openai.com |
| Google AI Studio (Google LLC) | Large language model inference | United States | ai.google.dev |
| Google Cloud Vertex AI (Google LLC) | Large language model inference and model hosting | United States | cloud.google.com/vertex-ai |
| Microsoft Azure AI Foundry (Microsoft Corporation) | Large language model inference | United States | azure.microsoft.com |
| Amazon Bedrock (Amazon Web Services, Inc.) | Large language model inference | United States | aws.amazon.com/bedrock |
| Moonshot AI | Large language model inference for AI-assisted design features | Singapore | moonshot.ai |
| Cerebras Systems, Inc. | High-throughput model inference | United States | cerebras.ai |
| Relace AI, Inc. | Code generation and editing model inference | United States | relace.ai |
| Fireworks AI, Inc. | Large language model inference | United States | fireworks.ai |
| Quiver | Vector and SVG generation model inference | United States | quiver.ai |
| OpenRouter, Inc. | Model routing and inference fallback. OpenRouter forwards a request to the model provider serving the selected model, which may be one of the providers listed above or another provider on its network | United States | openrouter.ai |
Web search and content retrieval
These providers receive the search query or web address when you ask an AI feature to search the web or read a page.
| Subprocessor | Purpose | Location | Website |
|---|---|---|---|
| Tavily | Web search for AI-assisted features | United States | tavily.com |
| Firecrawl (Mendable AI, Inc.) | Web page retrieval and conversion for AI-assisted features | United States | firecrawl.dev |
Product analytics and monitoring
| Subprocessor | Purpose | Location | Website |
|---|---|---|---|
| PostHog, Inc. | Product analytics and session replay. Session replay records on-screen activity in the product, which can include design canvas content | United States | posthog.com |
| Functional Software, Inc. (Sentry) | Error tracking and application performance monitoring | United States | sentry.io |
| Langfuse GmbH | AI observability. Records AI requests and responses — including prompt input, generated output, model, and latency — so we can debug, evaluate, and improve AI features | United States (Langfuse Cloud US region) | langfuse.com |
Customer communications
| Subprocessor | Purpose | Location | Website |
|---|---|---|---|
| Resend, Inc. | Transactional and marketing email delivery | United States | resend.com |
Internal operations
The following tools are used for internal operations and may process limited customer, applicant, or business contact information (for example, when a customer opens a support ticket, a candidate applies for a role, or a person is referenced in an internal document). They do not process customer product data.
| Subprocessor | Purpose | Location | Website |
|---|---|---|---|
| Google Workspace (Google LLC) | Email, calendar, and document collaboration | United States | workspace.google.com |
| Slack Technologies, LLC | Internal team communications | United States | slack.com |
| Linear Orbit, Inc. | Issue tracking and engineering operations | United States | linear.app |
| Notion Labs, Inc. | Internal documentation and knowledge base | United States | notion.so |
| GitHub, Inc. | Source code hosting and CI/CD | United States | github.com |
| Typeform (Typeform SL / Typeform US LLC) | Recruiting application forms and surveys | United States / European Union | typeform.com |
Notifying customers of changes
We will provide at least 30 days' notice before adding a new subprocessor that will process customer personal data. Customers with an active subscription may object to a new subprocessor during the notice period by writing to team@wonder.so; if we are unable to accommodate a reasonable objection, the customer may terminate the affected service.
Contact
Questions about this page or our processing of personal data can be sent to team@wonder.so.