Founding Platform/SRE Engineer

Founding Platform/SRE Engineer — Sandbox, Cost & Observability

Why this role

Forge's core promise is safe sandboxed agents + real‑time cost governance + auditable execution. You'll build the production foundations that make that promise true: secure ephemeral runtimes, token→USD metering, end‑to‑end observability, and safe deploy/rollback.

What you'll own

• Ephemeral execution environments
  for agents/tools (containers / Firecracker / WASM), with CPU/mem/disk/network quotas, secrets brokering, and isolation hardening

• Observability by default
  : OpenTelemetry traces/logs/metrics across IDE → API → orchestrator → agent runtime, plus SLIs/SLOs and alerting

• Cost governance infrastructure
  : accurate token→USD accounting, per‑tenant budgets, anomaly detection, enforcement hooks (throttle/downshift/queue)

• Release engineering
  : CI/CD, canaries/blue‑green, rollbacks, feature flags, backups/DR, ...