Deployed by our team
Our team sets up and configures your instance. Your IT team only grants access.
You can run the full superglue platform in your own cloud or data center. Your data, brain and playbooks are stored there. The superglue model, which is trained for implementation work, runs the agents. Our team deploys the instance and installs updates.
Our team sets up and configures your instance. Your IT team only grants access.
We install new versions of superglue on a schedule you approve.
Client data, credentials and company knowledge are stored in your own cloud or data center.
A self-hosted instance stores your data and your firm's implementation knowledge in your environment.
Client records that a migration moves, and the mappings and reconciliations built from them.
Your meeting notes, workspace documents and email, indexed in your environment.
Your standards, migration patterns and quality checks, stored with the rest of your data.
System credentials are stored in your environment, encrypted with a master key you own.
Model calls go to the superglue model on AWS Bedrock, which processes them in memory and stores nothing. There is no license check, and telemetry is off by default.
Self-hosted instances use the superglue model by default. We serve it through our model gateway. Your data, brain and playbooks stay stored in your environment.
superglue supplies the best model for each implementation task. The superglue model is trained specifically for implementation work.
Prompts are processed in memory on AWS Bedrock. Nothing is stored or logged, and prompts are never used for training. AWS is our only subprocessor.
New versions of the model reach your instance through our model gateway. You do not need to change your configuration.
Security teams often reject tools that store client data with a vendor. A self-hosted instance stores client data in your environment, which makes the review easier.
Firms that handle payment, health, public sector or cross-border personal data can run a single-tenant instance in the region they choose.
Client data is stored only in your environment. Model calls are processed in memory and nothing is stored. AWS is our only subprocessor.
You grant access and delegate a subdomain. We deploy and configure the instance and hand it over to you. Your team needs to do very little.
We create IAM roles with a superglue prefix. These roles can only change the resources we create.
For example superglue.yourcompany.com, pointed at the instance in your account.
We set up the runtime, storage, the superglue model and SSO with your identity provider. If you have on-prem systems, we also set up the Secure Gateway.
Your instance has the same features as our cloud. We install updates from then on.
We publish new versions of superglue and install them on your instance on a schedule you approve.
Our team runs every upgrade. You decide when it happens.
Every release is a versioned image. If a release causes a problem, we roll back to the previous version.
Monitoring and support are included in the enterprise plan.
The enterprise plan has the same features in our cloud and in your own environment.
| Feature | Our cloud | Your cloud or on-prem |
|---|---|---|
| Agents, Brain, Playbooks, Capabilities | ||
| Any API, database or legacy system | ||
| Scheduling & webhook triggers | ||
| Access rules (RBAC) & SSO | ||
| Run monitoring & Activity Tracker | ||
| Secure Gateway to on-prem systems | ||
| Data and brain stored in your environment | – | |
| Deployment & updates by superglue | On your schedule |
We set up a proof of concept with one of your implementation projects, deployed the way your security team requires.