Sign InOpen Brain
VercelEngineering PostOfficial Source

How Tailscale built a customer-facing model router on AI Gateway

Tailscale tied model access and ephemeral agent sandboxes to network identity, without issuing keys to agents. It is a concrete pattern for combining gateways, access control, and isolated execution.

Vercel · Sep 11, 2026
Open Source Open MarkdownOpen JSON
Source Summary

Aperture gives customers one API for **hundreds of models**, with access granted or revoked through tailnet identity. AI Gateway returns usage and cost per request, while Vercel Sandbox provides ephemeral agent execution without issuing a key to the agent.

Practical Implication

Builders should treat routing, identity, retention policy, and execution isolation as one system. The described flow validates identity before work begins, supports global or per-request **zero data retention**, and shuts the sandbox down afterward.

Agent-Ready Context
Aperture gives customers one API for **hundreds of models**, with access granted or revoked through tailnet identity. AI Gateway returns usage and cost per request, while Vercel Sandbox provides ephemeral agent execution without issuing a key to the agent.

Builders should treat routing, identity, retention policy, and execution isolation as one system. The described flow validates identity before work begins, supports global or per-request **zero data retention**, and shuts the sandbox down afterward.

This is a vendor case study rather than an independent security assessment. Tailscale reports moving from prototype to paying customers in **months**, but the material gives no benchmark for isolation, latency, or total cost against competing gateways and sandboxes.
Connected Context · Feed7 Judgment

This turns several separate infrastructure controls into one customer-facing operating pattern: identity is checked before routing, retention can be selected globally or per request, credentials stay outside the agent, and execution ends with sandbox shutdown. It reinforces external policy enforcement and isolation, while the vendor case study leaves comparative security, latency, and cost unvalidated.

Security Firewall for Agents — Ryan Dahl, DenoDeno’s external traffic firewall reinforces Aperture’s choice to keep credentials and enforcement outside the agent, while warning that the gateway holding those credentials becomes a concentrated trust point.Unlock Agent Autonomy: The Runtime for AI-Native Systems — Tushar Jain, DockerDocker’s task containment and externally judged capabilities provide the broader least-privilege rationale for Aperture’s identity-gated access and ephemeral execution.Claude Fable 5.1 now available on AI GatewayFable’s mandatory 30-day retention demonstrates why Aperture’s per-route retention policy affects model eligibility rather than being a separate compliance setting.From fork() to Fleet: Designing an Agent Sandbox Cloud — Abhishek Bhardwaj, OpenAIThe sandbox-cloud account adds stronger isolation, persistence, and recovery considerations beyond Aperture’s described ephemeral lifecycle, whose isolation properties are not benchmarked.
Context Map
infrasecuritycoding#gateways#sandboxing#agent-reliability
Uncertainty
This is a vendor case study rather than an independent security assessment. Tailscale reports moving from prototype to paying customers in **months**, but the material gives no benchmark for isolation, latency, or total cost against competing gateways and sandboxes.