Use case
AI coding for regulated industries.
Regulated teams have to explain where code and prompts go, who approved each change, and what stops runaway spend. Topsoil is built so each answer points at infrastructure you already operate.
What a reviewer will ask, and where the answer lives
| Question | Answer |
|---|---|
| Where do prompts and code go? | To the Amazon Bedrock, custom HTTPS or gateway endpoint you configure, in your own AWS account. No Thazen service, prompt store or telemetry. Data flow |
| Does the model provider train on it? | AWS states that Bedrock inputs and outputs are never shared with model providers or used to train base models. |
| Who approved this change? | A person: plan approval, diffs for every edit, and a pull request reviewed by a human. Topsoil does not merge. Controls |
| Who can use it? | The IAM or IAM Identity Center identities you grant, with a shipped least-privilege policy. Deployment |
| What stops spend? | Monthly user and account budgets before each model call, and an AWS budget action backstop. |
Topsoil does not make a system compliant with any framework. It keeps the evidence your program needs in the places your program already examines.
See it against your own codebase.
A pilot runs in your own AWS account, with your models and your material, and ends with findings you can act on.