Riza
Executes untrusted code generated by AI models in a secure, isolated runtime with sub-10ms cold starts, configurable environment variables and network access, and returns execution output.
An isolated runtime service that executes untrusted code generated by AI models, solving the problem of safely running LLM-generated code without risking production systems. It targets developers and engineering teams at AI-native companies, from startups to enterprises, who need to run code at scale. The service is API-first with SDKs for Python, TypeScript, and Go, and offers secure isolation with configurable environment variables and network access, plus sub-10ms cold starts. It can be deployed on the provider's cloud or self-hosted on-premises.
Key features
- Secure isolation and control
- Configurable environment variables and network access
- No cold starts (sub-10ms)
- API-first design with RESTful endpoints
- SDKs for Python, TypeScript, and Go
- Supports popular libraries like pandas and matplotlib
- Self-hosting option for on-premises deployment
- Horizontally scalable Kubernetes service
- No social media activity within the last 30 days
GTM channels
- Blog
- Community
- API
- Docs
- Changelog
ICP
- Software developers
- Engineering teams
- Enterprises