// AI Infrastructure Assessment

AI Stack
Audit Checklist

24 items. 6 layers. Find the gaps in your AI infrastructure.

Based on the 6-layer AI OS model from the Academy. Check off what you have, see your maturity level, and get direct links to the lessons that fill each gap. Progress saves automatically in your browser.

0/ 24Beginner

Beginner (0-8)Operator (9-16)Architect (17-24)

Hardware & Runtime

Agent Platform

Tool Layer (MCP)

Memory & State

Routing & Orchestration

Output & Automation

Hardware & Runtime

0/4

Dedicated compute for AI workloads (Mac/GPU/cloud)L7

Service manager configured (launchd/systemd/Docker)L7

Automatic restart on crashL7

Resource monitoring (CPU, memory, disk)L24

Agent Platform

0/4

Persistent agent running 24/7 (not just chat sessions)L9

Agent has file system accessL10

Agent can execute terminal commandsL10

CLAUDE.md or equivalent agent constitution definedL8

Tool Layer (MCP)

0/4

3+ MCP servers configuredL11

External API integrations (search, data, generation)L11

Structured error handling on tool failuresL15

Tool descriptions written for model comprehensionL11

Memory & State

0/4

Persistent memory across sessionsL19

Session resumption capabilityL19

Knowledge base / semantic searchL21

Context compounding (each session builds on the last)L20

Routing & Orchestration

0/4

Model routing by task type (not one model for everything)L25

Cost tracking per model / per taskL26

Provider failback chain (primary fails, fallback to secondary)L25

Multi-agent orchestration for complex tasksL30

Output & Automation

0/4

Cron-based pipelines (content, monitoring, reporting)L32

Git-based deployment automationL32

Output validation before deliveryL33

Human escalation path when confidence is lowL22

// The Code Whisperer

Fill the gaps

Every unchecked item links to the exact AI Academy lesson that teaches you how to implement it. 39 lessons from foundations to production architecture.

AI Academy