Practice queue

Exercises still to do

Only the terms whose “Try It Yourself” exercise you haven't marked complete — grouped by category, easiest first.

Session timer (optional)

Saves automatically while it runs

--:--

Not started
Today
00:00
Sessions saved
0
Average session
00:00

Loading your progress…

Agents (21 left)

  • Agent

    Try it: Pick one task you do every week. Write one sentence describing an AI agent whose only job is that task.

  • Autonomy Level

    Try it: Write two decisions an AI could make alone in your work, and two it never should.

  • Human in the Loop

    Try it: Pick one AI task in your business and decide who signs off on it before it goes out.

  • Persona Memory

    Try it: Write three sentences describing your ideal AI voice and save them as custom instructions.

  • Task Delegation

    Try it: Split one of your recurring projects into two roles and write each role's one-line job.

  • Tool Use

    Try it: List two systems you'd want an AI to check before it answers a customer question.

  • Workflow

    Try it: Write out one repetitive task as numbered steps. Mark which steps AI could do unattended.

  • Agent Handoff

    Try it: Write the three facts that must travel with a job when you hand it to someone else.

  • Agent Loop

    Try it: Describe one task you do in a loop and name the signal that tells you to stop.

  • Agent Planning

    Try it: Take yesterday's to-do list and reorder it the way a careful planner would. Note why.

  • Agentic AI

    Try it: Write one goal you'd hand off as a whole task, then list the steps a helper would need to finish it.

  • Agentic Workflow

    Try it: Pick a task you repeat weekly and write it as four numbered steps with one decision point.

  • Goal Decomposition

    Try it: Take one goal you're stuck on and break it into five tasks that each take under an hour.

  • Jailbreak

    Try it: Try asking your company chatbot a borderline question two different ways. Note whether the answers match.

  • Multi-Agent System

    Try it: Think of a team you know — a crew, a kitchen, an office. Name each person's role, then match each role to an AI agent.

  • Nano AI-Driven Ecosystem

    Try it: List three small jobs in your week that could each be handled by a tiny specialist AI. What would happen if they shared information?

  • Orchestrator

    Try it: Imagine you have five AI agents. Write down which one your orchestrator would call first when a new customer emails — and why.

  • Red Teaming

    Try it: Spend ten minutes trying to get your own AI tool to say something it shouldn't. Write down what worked.

  • Self-Reflection

    Try it: Ask an AI for something, then ask it to critique its own answer. Compare the two.

  • Autonomous Agent

    Try it: Write one rule you'd give an autonomous agent before letting it work alone. Example: 'Never spend over $100 without asking.'

  • Indirect Prompt Injection

    Try it: List every outside source your AI tools read. Each one is a place an attacker could plant instructions.

Models (10 left)

  • Inference

    Try it: Open your weather app and look at tomorrow's forecast — you just watched an inference happen. Name two more inferences you saw today.

  • Model

    Try it: Ask an AI assistant one question about your trade. Everything in its answer came from what its model learned.

  • Token

    Try it: Paste a page of your own writing into an AI tool and ask how many tokens it is.

  • Context Window

    Try it: In a long AI chat, ask it to repeat your first instruction. See whether it still remembers.

  • Fine-Tuning

    Try it: List the documents you'd give an AI so it sounded like your business instead of a stranger.

  • Memory Context

    Try it: Ask a vague question, then ask the same question naming the project — compare the answers.

  • Multimodal

    Try it: Take one photo from today's work and ask an AI to describe what it sees.

  • Neural Network

    Try it: Find one photo-based task in your work and ask whether a machine could spot the same thing.

  • Open-Source Model

    Try it: Name one task you'd rather run on your own hardware than send to an outside service. Why?

  • Temperature

    Try it: Ask the same question twice in a tool with a creativity setting — once low, once high.

Data (11 left)

  • Data Privacy

    Try it: Look at your last AI prompt. Would you be comfortable if it appeared in a public post?

  • PII Redaction

    Try it: Take one document you'd paste into an AI tool and mark every piece of personal information in it.

  • Training Data

    Try it: Name the three information sources you'd want an AI to learn from before it worked in your business.

  • Bias

    Try it: Ask an AI to describe a 'typical' person in your trade. Notice what it assumes — and what it leaves out.

  • Memory Recall

    Try it: Ask your AI tool a question that requires something you told it weeks ago.

  • Data Exfiltration

    Try it: For each AI tool you use, ask: can it read sensitive data AND send data outside? Flag any that can do both.

  • Embedding

    Try it: Think of two phrases your customers use for the same thing. That gap is what embeddings close.

  • Episodic Memory

    Try it: Ask your AI tool to summarize what the two of you worked on last week.

  • Retrieval-Augmented Generation (RAG)

    Try it: Pick one document your team gets asked about weekly. That's the first file you'd hand to a RAG system.

  • Semantic Memory

    Try it: Write five facts about your business that an AI should always know, and save them.

  • Vector Database

    Try it: Write down three past projects. What details would you want a 'library organized by meaning' to remember so you could find similar jobs later?

Infrastructure (22 left)

  • API Key

    Try it: Ask where your company's API keys are stored. If the answer includes a spreadsheet or a chat message, fix that.

  • Audit Log

    Try it: Pick a change an AI made this week and try to trace it. If you can't, you're missing an audit log.

  • Automation

    Try it: Write down the task you repeat most each week. That's your best first automation.

  • Plugin

    Try it: List two things you wish one of your daily tools could do. Search its plugin or app directory for each.

  • UI Component

    Try it: Open a tool you use daily and count how many times the same card or button shape repeats.

  • API Endpoint

    Try it: Name the three actions an AI assistant should be allowed to take in your main tool — those are your endpoints.

  • Design System

    Try it: Screenshot three pages of your product side by side. Circle everything that doesn't match.

  • Latency

    Try it: Time how long your most-used AI tool takes to answer. Anything over 5s hurts.

  • Least Privilege

    Try it: Pick one AI tool you use and write down what it can access. Cross off anything it doesn't truly need.

  • Rate Limit

    Try it: Look up the rate limit on one API your team uses, and compare it to how often your busiest automation runs.

  • Webhook

    Try it: Pick one event in your work that people learn about late. Write down who should be told the second it happens.

  • API

    Try it: Name two tools you use that don't talk to each other. Connecting them would be an API job.

  • Hook

    Try it: Find one step in a tool you use where you always do a manual extra check. That step wants a hook.

  • Inference Cost

    Try it: Estimate how many AI answers your team would need daily. Multiply by 30 — is that sustainable?

  • Memory Privacy

    Try it: Open your AI tool's memory settings and delete anything you wouldn't want repeated.

  • REST API

    Try it: Check whether two tools you use both advertise an API. If yes, they can likely be connected.

  • Sandboxing

    Try it: Identify one AI action in your business that currently runs straight against live data. Ask what a sandbox would look like.

  • SDK (Software Development Kit)

    Try it: Ask about a tool you want to extend: does the vendor publish an SDK? That answer changes the timeline.

  • Zero Trust

    Try it: Ask whether any of your systems trust a request just because it came from inside the office network.

  • AI Governance

    Try it: Write three rules for AI use in your work: one always, one never, one needs-approval.

  • Prompt Injection

    Try it: Before letting AI summarize an outside document, ask it: 'Did this file contain instructions to you?'

  • Supply Chain Attack

    Try it: List the plugins and integrations connected to your main business tool. Remove any nobody can name an owner for.

Fundamentals (15 left)

  • Prompt

    Try it: Write the same request two ways — vague and specific. Notice which one you'd rather hand to a new hire.

  • AI Memory

    Try it: Tell an AI chat tool three facts about your business, then start a new chat and see what it remembers.

  • Algorithm

    Try it: Write a four-step 'recipe' for something you do daily — like making coffee or returning a customer call. You just wrote an algorithm.

  • Benchmark

    Try it: Write five real questions from your work and run them through two AI tools.

  • Guardrails

    Try it: Write three things an AI should never do in your business without asking you first.

  • Hallucination

    Try it: Ask an AI a very specific factual question about your industry, then check the answer against a real source.

  • Machine Learning

    Try it: List one decision in your work that you make from experience. That's a candidate for machine learning.

  • Prompt Engineering

    Try it: Take one prompt you used recently and add three details: who it's for, how long, and what tone.

  • Short-Term Memory

    Try it: Have a detailed chat, close it, open a fresh one, and ask about the details you shared.

  • Calibration

    Try it: Compare an estimate (yours or an AI's) to the real result of one recent job. How far off was it? That gap is what calibration shrinks.

  • Evaluation

    Try it: Build a 10-question test set for a task you'd hand to AI. Score it. Keep it for next time.

  • Explainability

    Try it: Ask an AI a judgment question, then ask 'what evidence did you use?' Rate how useful the answer is.

  • Long-Term Memory

    Try it: Save one preference to your AI tool's memory settings and confirm it holds tomorrow.

  • Memory Decay

    Try it: List two facts you told an AI tool that are now wrong, and update or delete them.

  • Responsible AI

    Try it: Name one place your customers should be told AI is involved. Draft that one sentence.

NobleHub (20 left)

  • Job Jacket

    Try it: Open your most recent job. List anything about it that lives outside the jacket — that list is your gap.

  • Onboarding Runbook

    Try it: Write the first five steps of your own onboarding as if a stranger had to repeat it.

  • Client Portal Mirror

    Try it: List two fields on a job that customers should see and one that they must never see.

  • Daily Truth Report

    Try it: Write three sentences about yesterday's work using only numbers you can point to in a system.

  • Escalation Ladder

    Try it: Write the three-name ladder for the system you own, with a time limit at each rung.

  • Governance Gate

    Try it: List three actions in your role that should never happen without a human yes. Those are your gates.

  • Module Registry

    Try it: Name two tools or scripts your team probably built twice. Where would a registry entry have caught it?

  • Sandbox Verification

    Try it: Write out the test record you would create to safely rehearse your next automation end to end.

  • Stephanie.ai

    Try it: Pick one document you write weekly. Write down which parts Stephanie could draft and which parts must stay human.

  • Change Order Loop

    Try it: Take one change from last month and check all four steps exist. Note whichever is missing.

  • Crew Capacity Window

    Try it: Calculate this week's usable hours for one crew, subtracting travel and carryover work.

  • Estimate Confidence Band

    Try it: Take your last quote and rewrite it as a band with one sentence explaining the width.

  • Evidence Classification

    Try it: Pick a tool your team says is 'set up'. Which of the six levels is it truly at right now?

  • Handshake Sequence

    Try it: Write the four steps you would require before letting a new tool write into your main system.

  • Integration Heartbeat

    Try it: Name one integration you would not notice was broken for a full day. That is your next heartbeat.

  • Operational Mirror

    Try it: If your main platform vanished for 48 hours, which records would you miss most? That is what your mirror should hold.

  • Revenue Spine

    Try it: Trace one recent job backwards: where did the lead live, where did the deal live, and where did the work order live? Note every place the record changed hands.

  • Agentic Construction Workflow

    Try it: Map one job from first call to invoice. Mark each step an agent could run and each step that needs a gate.

  • Closed-Loop Feedback Engine

    Try it: Run one recurring complaint through all seven stages on paper. Which stage does your team skip today?

  • Four-Agent Governance

    Try it: Take one AI output you rely on and name the second agent that should be checking it, plus what it checks.