APEX
AI Proficient Executive. AI Company USA's coaching program for leaders who want to run their business with AI, on the tools they already use.
The APEXGlossary.
The words you will hear in every APEX session, defined in plain English for the person who owns the decisions.
38terms, defined plainly
05sections, from method to judgment
01page quick reference
Section A · 10 terms
Proficiency at its core
How an engagement runs, from the first signature to the final report.
Our approach
Your standard. Your pace.
APEX starts with your BASE-Eval, a look at how you use AI today, and works step by step toward the standard you set in your GOAL. Proficiency at its core means meeting that standard. Excellence comes from applying it consistently, by you and your organization, in the decisions you make. Along the way we keep data, money and trust protected, and anyone involved may call a stop at any time, without penalty. We measure progress together at the end with your REval.
Leader plus one
The core format: one leader plus one partner or operator who learns alongside them, so the capability stays inside the business.
Agreement
The signed start of an engagement. We confirm the fit first, and work begins only after the agreement is signed.
BASE-Eval
The baseline assessment taken at the start. It measures where you are with AI today, so progress is measured instead of guessed.
Weeks 1 to 2GOAL
Your own written statement of what success looks like, captured at intake in your words. Every session is steered toward it.
Weeks 1 to 2Session
A live two-hour working session on your real work and your own tools. The core engagement has twelve, about one a week.
Weeks 3 to 9Checkpoint
The midpoint review, around week six. We measure progress against your GOAL and adjust the plan.
Week 6REval
The reassessment at the end, weighted the same as the BASE-Eval so the two can be compared directly.
Weeks 10 to 11Progress report
The closing summary: where you started, where you said you wanted to be, and where you ended up.
Week 12Continuation
The month-to-month option after the core engagement, for leaders who want to keep building.
After week 12Section B · 10 terms
How the tools behave
What is actually happening when you type into an AI tool.
Large language model (LLM)
The kind of AI behind tools like Claude, ChatGPT and Gemini. It produces text from patterns learned in training, and it does not look things up unless it is given a tool to do so.
Model
A specific version of an AI system. Models trade speed, cost and depth of reasoning, so the right one depends on the job.
Prompt
The instructions you give the AI. Clear context, a defined task and an example of good output do more than clever wording.
Context
Everything the AI can see while it works: your prompt, attached files, earlier messages and any connected data. Better context, better answers.
Context window
The limit on how much the AI can hold in view at once. In very long conversations, early details can drop out of view.
Token
The unit AI tools use to count text, roughly three quarters of a word. Limits and pricing are often measured in tokens.
Hallucination
A confident answer that is false or made up: a wrong number, an invented source, a policy that does not exist.
Grounding
Tying the AI's answer to a source you trust, such as your own documents, so every claim can be traced and checked.
Reasoning model
A model that works through a problem in steps before it answers. Slower, and stronger on analysis, planning and math.
Custom instructions
Standing guidance the AI reads before every conversation: your role, your business, your standards. Set once, used every time.
Section C · 07 terms
Agents and automation
When AI stops answering and starts doing.
Agent
An AI that takes a series of steps toward a goal and uses tools along the way, instead of answering one message at a time.
Connector
A link that lets an AI tool read or act in another system, such as email, a calendar or a CRM. Every connector is an access decision.
MCP (Model Context Protocol)
An open standard for connecting AI tools to other software. Many connectors are built on it.
Workflow
A repeatable sequence of steps, some done by AI and some by people, that produces the same kind of result each time.
Human in the loop
A planned point where a person reviews, approves or corrects the AI's work before it moves on.
Stop point
A condition that makes an agent pause and hand back to a person, such as spending money, contacting a client or touching sensitive data.
Orchestration
Coordinating several AI agents or steps, each with a defined role, toward one outcome.
Section D · 08 terms
Leading with AI
The judgment that separates using AI from directing it.
AI proficiency
In APEX, the ability to identify, evaluate, direct and govern AI use for defined business decisions and workflows. Knowing when not to use AI is part of it.
Delegation
Deciding which work goes to AI, which stays with people and who checks the result. The core leadership skill in an AI-enabled business.
Verification
Checking AI output against the source before you rely on it. Quick to do, and costly to skip.
Value drift
When a system chases a number so well that the number replaces the purpose behind it. The subject of Pax Machina, Explained, Episode 3.
AI governance
The rules a business sets for how AI is used: which tools, which data, who approves what and how it is recorded.
Shadow AI
AI tools used at work without the business knowing or approving. Usually well meant, and a common source of data risk.
Data handling
Knowing what information may go into which tool. Client, financial and personal data need approved tools and clear rules.
Evaluation (eval)
A repeatable test that checks whether an AI system does a job well enough to be trusted with that job.
Section E · 03 terms
Judgment buckets
Three buckets for sorting any task before AI touches it.
Bucket A: AI drafts
Low-risk work AI can do first: summaries, first drafts, research outlines. You still read it before it leaves the building.
Bucket B: AI assists, you decide
AI prepares the options and the analysis. A person makes the call: pricing, hiring, vendor choices.
Bucket C: Owner only
Decisions that stay with a qualified person: legal, safety, financial commitments and anything that cannot be undone.
No term matches that.Try a shorter word, or clear the search.
Quick reference
Ten terms. One page.
BASE-EvalWhere you start, measured.
GOALSuccess, in your words.
REvalWhere you end, measured the same way.
PromptThe instructions you give the AI.
ContextEverything the AI can see while it works.
HallucinationA confident answer that is wrong.
AgentAI that takes steps, not just turns.
Human in the loopA person checks before it moves on.
VerificationCheck the source before you rely on it.
Value driftWhen the metric replaces the mission.
APEX coaching
Put the words to work.
APEX coaching turns these terms into working habits, inside your business and on the tools you already use.
Inquire about APEX Prefer email? taylor@aicompanyusa.com