Tap a track to see who it's for, what you'll practice in plain English, and the badge you earn.
Fundamental AI Execution Get reliable answers out of AI, every single time.
Who it's for
Anyone who already uses AI at work but gets inconsistent results—small business owners, coordinators, analysts and individual contributors.
Sales · Marketing · HR · Product · Engineering · Finance
What you'll practice
1.1 Context Design — Give AI the background it needs so it stops guessing what you meant. Aligned: NIST AI RMF 1.0 • OWASP LLM Top 10
1.2 Fact Auditing — Spot confident-sounding answers that are simply made up—before they reach a client. Aligned: RAG Triad Grounding Standards • NIST SP 800-218 (SSDF)
1.3 Data Analysis — Turn a messy spreadsheet export into a clear, decision-ready summary. Aligned: Structured Output Contract • JSON Schema Draft 2020-12
1.4 Data Privacy & Security Guardrails — Anonymize sensitive data (PII) and hardcode rules AI must never break. Anonymize personal identification numbers (SSNs, credit cards, emails, medical IDs) prior to payload transmission. Hardcode strict operational boundary rules (e.g. maximum discount caps, non-negotiable legal/compliance disclaimers). Block accidental leakage of confidential internal company data or strategy files. Aligned: OWASP LLM Top 10 • NIST SP 800-122
1.5 Token Compression — Get the same quality answer for a fraction of the cost. Aligned: Enterprise FinOps Framework • Asymmetric Token Pricing Spec
Shows you can brief AI clearly and catch its mistakes.
Agentic Workflows & Digital SOPs Let AI safely use your tools and follow your team's playbook.
Who it's for
Team leads and operators who want AI to actually do the work—drafting, filing, updating records—without handing over the keys.
Revenue ops · Recruiting ops · Product ops · Platform teams · Finance ops
What you'll practice
2.1 Tool Access Scoping — Give AI a read-only badge instead of full access to your inbox and files. Aligned: OWASP LLM Top 10 • Anthropic Model Context Protocol (MCP)
2.2 Team SOPs — Write your process down once so AI performs it the same way every time. Aligned: IEEE 1588 State Machine Standards • FSM Formalization
2.3 Execution Loops — Let AI retry and self-correct without spinning forever or burning budget. Aligned: ReAct (Reason + Act) Pattern • Microservice Circuit Breaker Standards
2.4 Human Approval Gates — Force a human sign-off before anything is sent, paid or published. Aligned: EU AI Act
2.5 Scheduled Routines — Put recurring work—weekly reports, pipeline hygiene—on autopilot. Aligned: SRE Scheduled Job Standards • Event-Driven Architecture (EDA)
Shows you can connect AI to real tools with the right guardrails.
Autonomous System Governance Run a team of AI workers with budgets, escalation and oversight.
Who it's for
Managers, directors and founders accountable for output, spend and risk when several AI workflows run at once.
Department heads · Ops leadership · Finance controllers · Founders
What you'll practice
3.1 Delegation & Handoffs — Split a big job across several AI workers without duplicated or lost work. Aligned: Multi-Agent Swarm Handoff Specs • Context Window Entropy Controls
3.2 Spending Caps — Cap daily AI spend and block charges nobody approved. Aligned: Enterprise Cloud FinOps • API Gateway Rate-Limiting Governance
3.3 Exception Triage — Review only the edge cases AI flagged, instead of every single item. Aligned: SRE Incident Management • Dead-Letter Queue (DLQ) Architecture
3.4 Quality Drift Monitoring — Notice when results quietly get worse—and fix the cause, not the symptom. Aligned: ISO/IEC 42001 • Continuous Eval Benchmarks
Shows you can govern AI at team scale—cost, quality and exceptions.