Author

Sofia Alvarez

Head of Engineering

Former Datadog principal engineer. Runs our agent runtime team. Has strong opinions about queues and observability.

Writing

Posts by Sofia Alvarez

2026-04-08

Why we built a sandboxed runtime for AI agents

Prompts are not a security boundary. The architecture we landed on after six months in production: V8 isolates, capability-scoped tools, kill-switches, observability.

2026-03-04

Why we chose multi-tenant isolation over single-tenant for security

Single-tenant feels safer. We argue, with the threat model in hand, that the opposite is true for agent workloads.

2026-02-09

Treating evals as production infrastructure

How we run 12,000 eval cases on every PR — and what we do with the cases that flake.

Posts by Sofia Alvarez

Why we built a sandboxed runtime for AI agents

Why we chose multi-tenant isolation over single-tenant for security

Treating evals as production infrastructure

Read more on the blog