Programs

Cursor, Grok, OpenAI, Anthropic.

We train teams to run a multi-model roster with pstack discipline instead of hoping a chat window replaces engineering.

01

Cursor + pstack

Install the engineering discipline: smallest change first, isolated state, verification before judgment, and the rule that green is not the same as safe to ship.

CursorpstackAgents
02

Grok

Use Grok for high-velocity planning, research, and agent loops without losing the review gate. We teach when to let it run and when a human has to sign.

GrokAgentsReview
03

OpenAI

Prompting, tool use, and evals that survive contact with production. We map model strengths onto real cloud and product work, not demo chat.

OpenAIPromptingEvals
04

Anthropic

Long-context reasoning, careful refactors, and specialist critics. We show how Claude fits a multi-model roster instead of becoming the whole company.

AnthropicClaudeReview

The Cloud Engineer Agent and IAM Identity Agent are the teaching artifacts. Teams see a real request become a reviewed pull request, then practice the same loop on their own stack.

Cloud Engineer workflowIAM Identity workflowBook a working session