Pulumi × Compostable AI
Pulumi x Compostable AI: AI agents manage infrastructure for 19 clients with two engineers
90 minutes to provision and deploy a new customer tenant to production
Engineer as craftsman to engineer as factory operator
Members read free
Unlock the full breakdown
Enter your email to open every case study in the library: the full screenshot, why it works, and the one thing worth stealing. One email unlocks all of them.
Free. No spam. Unsubscribe anytime.
You're in — enjoy the library.
Why it works
The story earns New Capability because the before-state was not merely slow provisioning but an architectural impossibility: infrastructure tooling that required human review was incompatible with an all-agent development model. The 90-minute tenant provisioning metric is concrete and attributable, and the mechanism (TypeScript type system enabling agent iteration, ESC hierarchical config, Neo via MCP server, policy packs for compliance) is explained in enough operational detail to be replicable. The philosophical quote from Dawson about shipping more code in two years than in fifteen adds human weight without being filler.
Steal this
The AGENTS.md file routing pattern: defining in a config file which agent handles which tool-specific task avoids the jack-of-all-trades problem and gives readers a concrete, immediately copyable architectural decision.
Click to enlarge ↗ This is editorial commentary and curation. The case study, screenshot, and all metrics are Pulumi's published work; we link to the source and lead with our analysis.