LLM (Large Language Model)
A neural network trained on massive text data to understand and generate language.
Why it matters when talking to AI: ChatGPT, Claude, Gemini — these are all LLMs. The engine behind every AI conversation.
The AI Nerdccionary
Speak Software — the vocabulary that unlocks AI capabilities.
You don't need to learn to code. But you need to speak software. When you say “cron” instead of “schedule something,” your AI agent doesn't guess — it creates a cron job. The right word = deterministic results.
Showing 92 of 92 terms
This glossary is organized by domain so you can learn the language of AI, software, infrastructure, APIs, security, and OpenClaw without getting lost in jargon.
🤖 AI & LLM Fundamentals
Token
The smallest unit of text an LLM processes — roughly a word or word fragment.
Why it matters when talking to AI: Everything costs tokens. Input tokens (what you send) + output tokens (what it generates) = your bill.
Context Window
The maximum amount of text an LLM can process at once.
Why it matters when talking to AI: 200K tokens sounds like a lot — until your agent has a long conversation and runs out of space.
Prompt
The text input you give an AI model to get a response.
Why it matters when talking to AI: The quality of your prompt directly determines the quality of the output. Garbage in, garbage out.
System Prompt
Hidden instructions that define how the AI behaves before you even start talking.
Why it matters when talking to AI: This is how you give an AI agent its personality, rules, and expertise. The user never sees it.
Temperature
Controls randomness in AI responses. Low = predictable, high = creative.
Why it matters when talking to AI: Set to 0 for code and facts. Set higher for creative writing. Most agents use low temperature.
Model
A specific trained AI system — like GPT-5, Claude Opus, or Gemini 3.
Why it matters when talking to AI: Models come in sizes: bigger = smarter but slower and more expensive. Pick the right one for the job.
Fine-tuning
Training a model further on your specific data to specialize it.
Why it matters when talking to AI: Want an AI that knows your industry? Fine-tuning is how you teach a general model your domain.
Embedding
Converting text into numbers (vectors) so computers can measure similarity.
Why it matters when talking to AI: How semantic search works — find things by meaning, not just keywords.
RAG (Retrieval-Augmented Generation)
Feeding an LLM relevant documents before it answers, so it uses real data instead of guessing.
Why it matters when talking to AI: The most common way to give AI access to your company's knowledge without fine-tuning the model.
Inference
Running a trained model to get predictions or responses. The "doing work" part.
Why it matters when talking to AI: Training builds the brain. Inference is the brain thinking. Every API call = inference = cost.
Hallucination
When AI confidently generates information that is factually wrong.
Why it matters when talking to AI: The #1 reason you can't blindly trust AI output. Always verify critical facts.
Agent
An AI that can use tools, make decisions, and take actions — not just chat.
Why it matters when talking to AI: The leap from chatbot to worker. Agents can browse the web, write files, run code, send emails.
Knowledge Graph
A database that stores facts as entities and relationships, not just text.
Why it matters when talking to AI: How TrueMem stores memory: "Acme Co" → client → Miami → roofing. Structure, not just words.
Vector Database
A database optimized for storing and searching embeddings (number representations of text).
Why it matters when talking to AI: Powers semantic search — find documents by meaning rather than exact keyword match.
Generative AI
AI that creates new content — text, images, code, video — rather than just analyzing existing data.
Why it matters when talking to AI: ChatGPT started the generative AI era. It doesn't just find answers, it writes them.
Reasoning AI
AI that can plan, reflect, decompose problems, and think step-by-step before answering.
Why it matters when talking to AI: O1/O3 made AI trustworthy by grounding responses in structured thinking. The second inflection.
Agentic AI
AI that can autonomously perform tasks, use tools, and iterate — not just respond to prompts.
Why it matters when talking to AI: Claude Code, Codex, Cursor — all agentic. The third inflection: AI that does, not just talks.
Frontier Model
The most capable AI model available at any given time, pushing the boundary of what's possible.
Why it matters when talking to AI: GPT-5, Claude Opus 4, Gemini 3 — the frontier keeps moving. Today's frontier is tomorrow's baseline.
Open Source / Open Weights
Models whose code or weights are publicly available for anyone to use, modify, and deploy.
Why it matters when talking to AI: Llama, Mistral, NemoTron — open models let you run AI locally without paying per-token API costs.
Scaling Law
The principle that AI performance improves predictably with more data, compute, and parameters.
Why it matters when talking to AI: Why companies keep building bigger. More compute = better AI. The flywheel hasn't stopped yet.
Pre-training
The initial training phase where a model learns from massive datasets (the internet, books, code).
Why it matters when talking to AI: This is where the model builds its general knowledge. Costs millions of dollars and months of time.
Post-training
Training after pre-training that refines the model for specific tasks — instruction following, safety, reasoning.
Why it matters when talking to AI: This is where models get aligned with human preferences. RLHF, DPO, and other techniques live here.
Benchmark
A standardized test to measure and compare AI model performance.
Why it matters when talking to AI: SWE-bench, CORE, MMLU — numbers that tell you how smart a model is on specific tasks. Useful but limited.
AGI (Artificial General Intelligence)
AI that can think, learn, and adapt across many different tasks like humans — not yet achieved.
Why it matters when talking to AI: The north star everyone is building toward. When someone says "we're close to AGI" — they're speculating.
Attention Mechanism
How AI models focus on the most relevant parts of input when generating a response.
Why it matters when talking to AI: The breakthrough that made transformers work. It's why GPT can connect the beginning of a paragraph to the end.
Deep Learning
Machine learning using neural networks with many layers to learn complex patterns from data.
Why it matters when talking to AI: The technology under every modern AI. When someone says "AI" in 2026, they mean deep learning.
Foundation Model
A large AI model trained on broad data that serves as the base for building specific applications.
Why it matters when talking to AI: GPT-5, Claude Opus, Gemini — these are foundation models. You fine-tune them for your domain.
Multimodal
AI that can process multiple types of data — text, images, audio, video — simultaneously.
Why it matters when talking to AI: GPT-4o can see images and hear audio. Multimodal = the AI isn't limited to just text.
Neural Network
A computing system inspired by the brain, using layers of connected nodes to process information.
Why it matters when talking to AI: The architecture under all modern AI. Every LLM is a neural network — a very, very large one.
Parameter
An adjustable value inside a model that gets tuned during training. More parameters = more capacity to learn.
Why it matters when talking to AI: GPT-4 has ~1.8 trillion parameters. Bigger models learn more nuance but cost more to run.
RLHF (Reinforcement Learning from Human Feedback)
A training method where humans rate AI responses to teach the model what good output looks like.
Why it matters when talking to AI: How ChatGPT learned to be helpful instead of just predicting text. Human feedback shapes behavior.
Training
The process of showing an AI model millions of examples so it learns patterns and relationships.
Why it matters when talking to AI: Training costs millions of dollars and months of compute. Once trained, the model is used for inference.
Machine Learning
Algorithms that improve performance by learning from data — without being explicitly programmed for each task.
Why it matters when talking to AI: The broader field that includes deep learning and AI. Most "AI" is actually machine learning under the hood.
Conversational AI
AI systems designed to understand and generate human-like conversation — chatbots, voice assistants, agents.
Why it matters when talking to AI: ChatGPT, Claude, Gemini chat — all conversational AI. The interface most people interact with.
Cost Per Token
How much it costs to generate one token of AI output.
Why it matters when talking to AI: GPT-4.1-mini: $0.40/M input. Opus: $15/M input. The model you pick determines your operating cost.
Token Budget
An allocation of compute resources measured in tokens rather than dollars.
Why it matters when talking to AI: Future employees will get token budgets alongside salary. More tokens = more AI amplification.
Throughput
The volume of tokens an AI system can produce in a given time at a given power budget.
Why it matters when talking to AI: High throughput = serve more users. The vertical axis on every AI factory performance chart.
Token Economics
The cost model where AI services price based on tokens consumed — input tokens + output tokens = your bill.
Why it matters when talking to AI: OpenAI charges per million tokens. Understanding token economics is understanding your AI operating cost.
Time to First Token (TTFT)
How long between sending a prompt and receiving the first token of the response.
Why it matters when talking to AI: Low TTFT = the AI feels responsive. High TTFT = the user waits and gets frustrated. UX-critical metric.
🧠 AI Agent Architecture
Harness
Everything around the model that determines how it works — tools, memory, verification, execution environment.
Why it matters when talking to AI: Same model, different harness, nearly double the performance. The harness matters more than the model.
MCP (Model Context Protocol)
An open standard for connecting AI agents to external tools and data sources.
Why it matters when talking to AI: Backed by Anthropic, OpenAI, Google, Microsoft. The USB port for AI agents.
Plugin
A module that extends an agent's capabilities — memory, search, tools, integrations.
Why it matters when talking to AI: LCM, TrueMem, QMD — these are all plugins that give an agent abilities it doesn't have natively.
Skill
A reusable instruction set that teaches an agent how to perform a specific task.
Why it matters when talking to AI: Like a recipe card for AI. Install a skill, and the agent knows how to do that job.
Sandbox
An isolated environment where code runs without access to the rest of the system.
Why it matters when talking to AI: Codex runs in sandboxes — safe but limited. Claude Code runs on your machine — powerful but riskier.
Lock-in
When switching tools requires rebuilding all your workflows, not just learning new commands.
Why it matters when talking to AI: Your team builds processes around a harness. Switching resets everything to zero. Price this into decisions.
Context Management
Systems that control what information stays in the AI's working memory and what gets compressed.
Why it matters when talking to AI: Without this, long conversations lose old context. LCM solves this with lossless compaction.
Session
A continuous conversation between you and an agent, with shared context.
Why it matters when talking to AI: Sessions can persist across restarts if the agent has memory. Without memory, every session starts fresh.
Auto-recall
Automatically searching memory and injecting relevant facts before the AI responds.
Why it matters when talking to AI: You ask about a client — the agent silently searches its graph and already knows the context.
Librarian
A process that extracts structured facts from conversations and writes them to the knowledge graph.
Why it matters when talking to AI: How experiences become permanent memory. Conversations → facts → graph → recall.
Entity Resolution
Recognizing that different names refer to the same thing.
Why it matters when talking to AI: "Acme Co" and "Acme Roofing Co" are the same client — the graph knows that.
Temporal Invalidation
Marking facts as expired when they're no longer true.
Why it matters when talking to AI: Client moved offices? The old address gets invalidated, not deleted. Time-aware memory.
Sub-agent
An agent spawned by another agent to handle a specific task.
Why it matters when talking to AI: How complex work gets distributed — your SEO agent, QC agent, and dev agent are sub-agents.
Gateway
The central process that routes messages, manages sessions, and loads plugins for an agent platform.
Why it matters when talking to AI: The traffic controller. Every message flows through the gateway to reach the right agent.
Workspace
The directory where an agent stores its files, memory, skills, and configuration.
Why it matters when talking to AI: The agent's home base. Everything it knows and can do lives here.
Cron Job
A scheduled task that runs automatically at set intervals.
Why it matters when talking to AI: Run memory cleanup every Sunday. Check client rankings every Monday. Automation without prompting.
Tool Calling / Function Calling
When an AI model requests to execute a specific tool or function during its response.
Why it matters when talking to AI: The agent decides it needs to search the web, read a file, or query a database — and does it mid-conversation.
Orchestrator
The component that coordinates multiple AI agents, decides task routing, and manages workflow.
Why it matters when talking to AI: Like a project manager for AI agents. It decides who does what and in what order.
Agent Runtime
The execution environment that manages an AI agent's lifecycle — startup, state, tools, and shutdown.
Why it matters when talking to AI: OpenClaw's gateway is an agent runtime. It keeps the agent alive, routes messages, loads plugins.
Verification Loop
An automated check that validates AI output against criteria before proceeding.
Why it matters when talking to AI: The agent writes code → runs tests → checks results → iterates. Without verification, output is unreliable.
Guardrails
Safety mechanisms that prevent AI agents from taking harmful or unintended actions.
Why it matters when talking to AI: Content filters, permission gates, scope limits — the rules that keep an agent from going off the rails.
Multi-Agent System
Architecture where multiple specialized AI agents collaborate to solve complex problems.
Why it matters when talking to AI: One agent researches, another writes, a third reviews. Division of labor, but for AI.
Structured Output
Forcing the AI to respond in a specific format — JSON, tables, lists — instead of free-form text.
Why it matters when talking to AI: When you need data, not prose. JSON mode, schema enforcement, and format instructions turn AI into a reliable data tool.
🌐 Web & Architecture
HTML
The markup language that structures content on every web page.
Why it matters when talking to AI: Every website is HTML at its core. AI agents generate it, read it, and manipulate it.
CSS
The styling language that controls how web pages look — colors, layout, fonts.
Why it matters when talking to AI: Separate from HTML. Change the CSS, change the look without touching the content.
JavaScript
The programming language that makes websites interactive and dynamic.
Why it matters when talking to AI: Runs in browsers, on servers (Node.js), and powers most AI agent interfaces.
Server
A computer that provides services to other computers over a network.
Why it matters when talking to AI: Your AI agent runs on a server — either your Mac Mini at home or a VPS in the cloud.
HTTP/HTTPS
The protocol browsers use to communicate with servers. HTTPS adds encryption.
Why it matters when talking to AI: Every API call, every web page load, every AI request travels over HTTP/HTTPS.
REST API
An architectural style for building web APIs using standard HTTP methods (GET, POST, PUT, DELETE).
Why it matters when talking to AI: Most AI services expose REST APIs. You send a request, you get JSON back.
JSON
A lightweight data format used everywhere for API communication and configuration.
Why it matters when talking to AI: Every API response, every config file, every AI tool call uses JSON. Learn to read it.
Normalize / Normalization
Organizing data to reduce redundancy and ensure consistency — making sure the same thing is always represented the same way.
Why it matters when talking to AI: If your data calls one client "Acme" and another record says "ACME Inc" — that's not normalized. AI can't connect what it can't match.
Schema
The structure that defines how data is organized — what fields exist, what types they are, how they relate.
Why it matters when talking to AI: A knowledge graph has a schema: entity types, relationship types, required fields. Without a schema, data is chaos.
Query
A request for specific data from a database or search system.
Why it matters when talking to AI: Every time your agent searches memory, it runs a query. Better queries = better recall = smarter responses.
Database
An organized collection of data stored and accessed electronically.
Why it matters when talking to AI: SQL databases store structured data. Vector databases store embeddings. Graph databases store relationships. AI uses all three.
Index
A data structure that makes searching faster — like a book's index lets you find topics without reading every page.
Why it matters when talking to AI: Without indexes, every search scans everything. With indexes, your agent finds facts in milliseconds instead of seconds.
Structured Data
Data organized in a predefined format — rows, columns, tables, fields with types.
Why it matters when talking to AI: Spreadsheets, databases, JSON — structured data is what AI can reliably process. The ground truth of business.
Unstructured Data
Data without a predefined format — PDFs, emails, images, videos, conversations.
Why it matters when talking to AI: 90% of the world's data is unstructured. AI finally makes it searchable and useful through embeddings.
🛠️ Infrastructure & DevOps
VPS (Virtual Private Server)
A virtual machine you rent in the cloud — your own server without owning hardware.
Why it matters when talking to AI: The most common way to host an AI agent 24/7 without running a computer at home.
SSH (Secure Shell)
A protocol for securely connecting to and controlling remote servers.
Why it matters when talking to AI: How you access your VPS. Type commands on your laptop, they execute on the server.
CLI (Command Line Interface)
A text-based interface for interacting with software by typing commands.
Why it matters when talking to AI: Most AI tools are CLI-first. If you can't use a terminal, you're locked out of the best tools.
Terminal / Shell
The program where you type CLI commands. Bash, Zsh, PowerShell are common shells.
Why it matters when talking to AI: Your AI agent lives in the terminal. Learning basic terminal commands unlocks everything.
Bash
The default shell on Linux/Mac. A scripting language for automating system tasks.
Why it matters when talking to AI: Claude Code's philosophy: "bash is all you need." Composable Unix primitives replace specialized tools.
Docker
A platform for running applications in isolated containers with all their dependencies packaged together.
Why it matters when talking to AI: Run Neo4j, Graphiti, and your AI stack in containers — reproducible and portable.
Container
A lightweight, isolated environment that packages an application with everything it needs to run.
Why it matters when talking to AI: Like a shipping container for software. Same contents, runs the same way everywhere.
Daemon / Service
A program that runs continuously in the background — like your AI agent's gateway.
Why it matters when talking to AI: systemd, launchd — these keep your agent alive even when you're not looking.
Port
A number that identifies a specific service on a server (e.g., port 443 for HTTPS, port 3000 for your app).
Why it matters when talking to AI: Your AI gateway runs on a port. If two things try to use the same port, neither works.
Tunnel
A secure connection that makes a local service accessible from the internet.
Why it matters when talking to AI: Cloudflare Tunnel, Tailscale — expose your home AI agent to the world without opening firewall ports.
Environment Variable (.env)
A configuration value stored outside the code — usually secrets like API keys.
Why it matters when talking to AI: Never hardcode passwords. Store them in .env files. Your agent reads them at startup.
Firewall
A security system that controls what network traffic is allowed in and out.
Why it matters when talking to AI: Your server's bouncer. Only let in SSH (port 22) and your agent's traffic. Block everything else.
Cron
A time-based job scheduler built into Linux/Mac.
Why it matters when talking to AI: Run cleanup at midnight. Check rankings every Monday. The engine behind scheduled agent tasks.
Linux
A free, open-source operating system that runs most of the world's servers, cloud infrastructure, and AI systems.
Why it matters when talking to AI: Your VPS almost certainly runs Linux. Most AI tools assume Linux. If you're on a Mac, you're already close — macOS is Unix-based.
Ubuntu / Debian
Popular Linux distributions. Ubuntu is based on Debian and is the most common choice for servers and AI workloads.
Why it matters when talking to AI: When a tutorial says "install on Ubuntu" — that's the operating system. Most VPS providers default to Ubuntu.
sudo
A command that runs another command with administrator (root) privileges. Short for "superuser do."
Why it matters when talking to AI: When you see "sudo apt install..." — sudo is asking for permission to install system-wide. It's the "please" of Linux.
apt / brew
Package managers that install software. apt for Linux (Ubuntu/Debian), brew (Homebrew) for macOS.
Why it matters when talking to AI: "apt install nodejs" on Linux, "brew install node" on Mac. These are how you install the tools your AI agent needs.
Root
The all-powerful administrator account on Linux. Has unrestricted access to everything on the system.
Why it matters when talking to AI: Running as root is dangerous — one wrong command can destroy the system. That's why sudo exists: temporary root access.
chmod / Permissions
Linux commands that control who can read, write, or execute a file.
Why it matters when talking to AI: "chmod +x script.sh" makes a script executable. Without it, the system won't run your script even if the code is correct.
grep
A command that searches for text patterns inside files. One of the most-used Linux tools.
Why it matters when talking to AI: Looking for an error? "grep error logfile.txt" finds every line containing "error." Your AI agent uses grep constantly.
curl / wget
Commands that download files or make HTTP requests from the terminal.
Why it matters when talking to AI: Test an API? "curl https://api.example.com". Download a file? "wget https://...". Essential for debugging AI services.
pipe ( | )
Sends the output of one command as input to another. The vertical bar character.
Why it matters when talking to AI: "cat log.txt | grep error | wc -l" — read the file, filter for errors, count them. Unix philosophy: small tools chained together.
nano / vim
Text editors that run in the terminal. nano is simple, vim is powerful but has a steep learning curve.
Why it matters when talking to AI: Need to edit a config file on your server? "nano openclaw.json" opens it right there. Works on Mac and Linux.
cat
Displays the contents of a file in the terminal. Short for "concatenate."
Why it matters when talking to AI: "cat .env" shows your environment variables. First thing you do when debugging.
ls
Lists files and folders in the current directory.
Why it matters when talking to AI: "ls -la" shows everything including hidden files (like .env). The most-typed command in any terminal.
cd
Change directory — move to a different folder.
Why it matters when talking to AI: "cd ~/.openclaw" takes you to the OpenClaw config folder. "cd .." goes up one level.
mkdir / rm / cp / mv
Basic file operations: make directory, remove, copy, move/rename.
Why it matters when talking to AI: mkdir creates folders. rm deletes (careful — no undo). cp copies. mv moves or renames.
tail / head
Show the last (tail) or first (head) lines of a file.
Why it matters when talking to AI: "tail -f log.txt" follows a log file in real-time — essential for watching your AI agent work.
🧬 Development & Version Control
Git
A version control system that tracks changes to files over time.
Why it matters when talking to AI: Every AI project uses Git. It's how you save, share, and undo changes to code and config.
Repository (Repo)
A project folder tracked by Git, containing all files, history, and branches.
Why it matters when talking to AI: Your AI agent's workspace is a repo. Codex treats the repo as the single source of truth.
Commit
A saved snapshot of your changes with a description of what changed.
Why it matters when talking to AI: Like a save point in a game. You can always go back to any previous commit.
Push / Pull
Push sends your local commits to a remote server. Pull downloads others' commits to your machine.
Why it matters when talking to AI: Collaborate with AI: it pushes code, you pull and review. Standard Git workflow.
Branch
A parallel version of your code where you can make changes without affecting the main version.
Why it matters when talking to AI: AI agents work on branches. When the work is done, it gets merged back to main.
Merge / Pull Request (PR)
Combining changes from one branch into another, usually with review.
Why it matters when talking to AI: Codex's entire workflow: agent works on a branch, submits a PR, human reviews and merges.
Clone
Creating a local copy of a remote repository.
Why it matters when talking to AI: First step to working with any project: clone it, then start making changes.
.gitignore
A file that tells Git which files to exclude from tracking — like secrets and build artifacts.
Why it matters when talking to AI: Never commit .env files, API keys, or node_modules. .gitignore keeps them out.
npm / pip
Package managers for JavaScript (npm) and Python (pip) — install and manage dependencies.
Why it matters when talking to AI: Most AI tools install via npm or pip. One command pulls in everything you need.
Dependency
External code your project relies on to function.
Why it matters when talking to AI: Your AI agent depends on dozens of packages. Keep them updated, or things break silently.
Linter
A tool that checks code for errors, style issues, and potential bugs before you run it.
Why it matters when talking to AI: OpenAI discovered linter errors double as remediation instructions — tell the AI what to fix.
CI/CD (Continuous Integration / Deployment)
Automated systems that test and deploy code every time changes are pushed.
Why it matters when talking to AI: Push code → tests run automatically → if they pass, deploy. No manual steps.
🔌 APIs & Integration
API (Application Programming Interface)
A defined way for two systems to communicate with each other.
Why it matters when talking to AI: Every AI service is accessed through an API. Your agent calls APIs to think, search, and act.
API Key
A secret string that identifies and authenticates you when calling an API.
Why it matters when talking to AI: Like a password for machines. Keep it secret. Rotate it regularly. Never commit it to Git.
Endpoint
A specific URL where an API receives requests.
Why it matters when talking to AI: https://api.openai.com/v1/chat/completions — that's the endpoint your agent calls to think.
Webhook
A URL that receives automatic notifications when something happens in another system.
Why it matters when talking to AI: Get notified instantly when a form is submitted, an email arrives, or a build finishes.
Rate Limit
A cap on how many API calls you can make per minute/hour/day.
Why it matters when talking to AI: Hit the limit and your agent stops working until it resets. Budget your calls wisely.
OAuth
An authentication standard that lets one service access another on your behalf without sharing your password.
Why it matters when talking to AI: How Google, GitHub, and other services let AI agents act as you — securely.
Payload
The data sent in an API request or response — usually JSON.
Why it matters when talking to AI: Your prompt is the payload you send. The AI's response is the payload you receive.
Model Context Protocol (MCP)
An open standard for connecting AI agents to external tools and data sources.
Why it matters when talking to AI: The "USB port" for AI. Any tool that speaks MCP works with any agent that speaks MCP.
🔒 Security
API Key Security
Practices for protecting API keys — never expose in code, use .env files, rotate regularly.
Why it matters when talking to AI: One leaked API key can cost thousands in unauthorized usage. Treat keys like passwords.
.env File
A file that stores environment variables (secrets, config) outside of your code.
Why it matters when talking to AI: The standard way to separate secrets from code. Always in .gitignore.
SSH Key
A cryptographic key pair used for secure server authentication without passwords.
Why it matters when talking to AI: More secure than passwords. Your private key stays on your machine, public key goes on the server.
Encryption
Converting data into an unreadable format that can only be decoded with a key.
Why it matters when talking to AI: HTTPS encrypts web traffic. SSH encrypts terminal sessions. Without encryption, everything is visible.
Sandbox
An isolated environment where code runs without access to the rest of the system.
Why it matters when talking to AI: Codex runs agents in sandboxes — they can't touch your real files. Safe but limited.
Permissions
Rules that control what a user or program can read, write, or execute.
Why it matters when talking to AI: Your AI agent needs carefully scoped permissions. Too broad = security risk. Too narrow = can't work.
Confidential Computing
A security model where even the server operator can't see your data or models during processing.
Why it matters when talking to AI: Deploy AI on any cloud without trusting the cloud provider. Your data stays encrypted even in use.
Sovereign AI
AI infrastructure owned and operated within a country's borders under local laws.
Why it matters when talking to AI: Every nation wants AI that respects their data laws. Sovereign AI = local models + local compute.
Zero Trust
A security model where nothing is trusted by default — every request must be verified.
Why it matters when talking to AI: No more "inside the firewall = safe." Every AI agent call gets checked. Every time.
🧰 Developer Tools
Chrome DevTools
Built-in browser tools for inspecting web pages, debugging JavaScript, and analyzing performance.
Why it matters when talking to AI: Essential for debugging AI-generated web pages and checking API calls in the Network tab.
Console
A panel in DevTools or terminal where log messages and errors appear.
Why it matters when talking to AI: When something breaks, the console tells you why. First place to look.
Logs
Recorded messages from a running system — errors, events, status updates.
Why it matters when talking to AI: Your AI agent writes logs. When it misbehaves, the logs tell the story.
Debugging
The process of finding and fixing errors in code or configuration.
Why it matters when talking to AI: Most AI agent problems are config errors, not code bugs. Check logs, check .env, check permissions.
Playwright / Puppeteer
Tools for automating web browsers — clicking, typing, screenshotting programmatically.
Why it matters when talking to AI: How AI agents interact with websites. Playwright can test, scrape, and validate visual output.
IDE (Integrated Development Environment)
A code editor with built-in tools — VS Code, Cursor, JetBrains.
Why it matters when talking to AI: Where you write and edit code. Cursor is an IDE with AI built in. VS Code is the most popular.
Testing
Verifying that code works correctly through automated checks.
Why it matters when talking to AI: AI-generated code needs testing just like human code. Untested code is untrustworthy code.
✏️ Prompt & Context Engineering
Prompt Engineering
The practice of crafting inputs to get better, more reliable outputs from AI models.
Why it matters when talking to AI: Not just "asking nicely" — it's structuring context, examples, and constraints for predictable results.
Context Engineering
The discipline of controlling what information surrounds a prompt — memory, documents, examples, instructions.
Why it matters when talking to AI: Prompt engineering is what to ask. Context engineering is what information to supply so the answer is reliable.
Chain of Thought (CoT)
Asking the AI to show its reasoning step-by-step before giving a final answer.
Why it matters when talking to AI: Adding "think step by step" dramatically improves accuracy on complex problems. Research-proven since 2022.
Few-Shot Prompting
Giving the AI a few examples of the desired input/output format before asking it to perform the task.
Why it matters when talking to AI: Show 2-3 examples of what you want, and it copies the pattern. Much more reliable than just describing it.
Zero-Shot Prompting
Asking the AI to perform a task with no examples — just instructions.
Why it matters when talking to AI: Most casual AI use is zero-shot. Works for simple tasks, but complex ones need examples or reasoning chains.
Tree of Thought (ToT)
AI explores multiple reasoning paths simultaneously and picks the best one.
Why it matters when talking to AI: Like brainstorming three approaches before committing. Better for planning than single-path thinking.
ReAct (Reasoning + Acting)
A pattern where the AI alternates between thinking about what to do and actually doing it with tools.
Why it matters when talking to AI: Think → Act → Observe → Think → Act. This is how most AI agents work internally.
Persona Prompting
Giving the AI a specific role or identity to shape its responses.
Why it matters when talking to AI: "You are a senior SEO consultant with 15 years experience" — the persona shapes how the AI approaches problems.
Structured Output
Forcing the AI to respond in a specific format — JSON, tables, lists — instead of free-form text.
Why it matters when talking to AI: When you need data, not prose. JSON mode and schema enforcement turn AI into a reliable data tool.
Prompt Injection
A security attack where malicious text tricks an AI into ignoring its instructions.
Why it matters when talking to AI: Someone hides "ignore all previous instructions" in a document your agent reads. Real threat, no perfect defense yet.
Jailbreak
Techniques that bypass an AI model's safety restrictions to produce forbidden content.
Why it matters when talking to AI: Every model has guardrails. Jailbreaks try to get around them. Important to understand for security.
Vibe Coding
Writing software by describing what you want in natural language and letting AI generate the code.
Why it matters when talking to AI: You describe the vibe, the AI writes the code. Controversial but increasingly common since 2025.
System Prompt
Hidden instructions that define how the AI behaves before the user starts talking.
Why it matters when talking to AI: This is where you set the AI's personality, rules, and expertise. The user never sees it.
Prompt Chaining
Breaking a complex task into a sequence of smaller prompts, where each output feeds into the next.
Why it matters when talking to AI: Better than one giant prompt. Each step is focused, verifiable, and easier to debug when something goes wrong.
Self-Consistency
Running the same prompt multiple times and picking the most common answer for higher reliability.
Why it matters when talking to AI: Like asking three doctors instead of one. If they all agree, you're more confident in the answer.
One-Shot Prompting
Giving the AI exactly one example before asking it to perform the task.
Why it matters when talking to AI: The middle ground between zero-shot (no examples) and few-shot (multiple examples). Often enough for simple formatting tasks.
Top-p (Nucleus Sampling)
A parameter that controls word diversity by only considering the top percentage of likely next words.
Why it matters when talking to AI: Temperature controls randomness. Top-p controls vocabulary breadth. Both shape how creative or predictable the output is.
Meta Prompting
Prompting that focuses on the structure and logic of how to approach a problem, rather than giving specific examples.
Why it matters when talking to AI: Instead of showing the AI examples, you describe the thinking pattern it should follow. Higher-level than few-shot.
Prompt Template
A reusable prompt structure with fixed sections — role, context, constraints, output format — that you fill in per task.
Why it matters when talking to AI: Like a form for AI. COSTAR (Context, Objective, Style, Tone, Audience, Response) is a popular template framework.
Role Prompting
Assigning a specific job title or expertise area to the AI before giving it a task.
Why it matters when talking to AI: "You are a senior marketing strategist" vs "You are a data analyst" — different roles produce different approaches to the same question.
✍️ Copywriting Frameworks
AIDA
Attention → Interest → Desire → Action. The most widely used copywriting framework.
Why it matters when talking to AI: Capture attention with the headline, build interest, create desire for the solution, then prompt the action. Works for ads, emails, landing pages.
PAS
Problem → Agitate → Solution. Identify the problem, amplify the pain, then present the fix.
Why it matters when talking to AI: The framework behind every late-night infomercial and every effective sales page. Name the pain before offering the cure.
BAB
Before → After → Bridge. Show the current state, the desired state, and your product as the bridge.
Why it matters when talking to AI: Paint the 'before' picture (struggling), the 'after' picture (thriving), then position your service as the bridge between them.
FAB
Features → Advantages → Benefits. List what it does, why that's better, and what the customer gets.
Why it matters when talking to AI: Features are what you built. Benefits are what the customer feels. FAB connects the two.
4Ps (Marketing Mix)
Price, Product, Promotion, Place — the four fundamental elements of bringing a product to market.
Why it matters when talking to AI: Every marketing strategy starts here. What are you selling, at what price, how are you promoting it, and where?
📈 Marketing & SEO
SEO (Search Engine Optimization)
The practice of improving a website's visibility in organic, unpaid search results.
Why it matters when talking to AI: AI agents can do SEO research, write optimized content, and monitor rankings — but they need to understand what SEO actually measures.
SERP (Search Engine Results Page)
The page displayed by a search engine after a user enters a query.
Why it matters when talking to AI: Understanding SERPs is understanding where your clients appear. AI tools analyze SERPs to find opportunities.
E-E-A-T
Experience, Expertise, Authoritativeness, Trustworthiness — Google's content quality framework.
Why it matters when talking to AI: Google's quality bar. AI-generated content needs to demonstrate real E-E-A-T or it ranks poorly.
Backlink
A link from another website pointing to yours — a key signal for search engine rankings.
Why it matters when talking to AI: Link building is one of the hardest SEO tasks. AI can prospect and draft outreach but can't replace relationships.
Organic Traffic
Visitors who find your website through unpaid search results, not ads.
Why it matters when talking to AI: The metric every SEO client cares about. More organic traffic = less ad spend needed.
Local SEO
Optimizing a business to appear in location-based search results and Google Maps.
Why it matters when talking to AI: Most agency clients are local businesses. Local SEO is where the money is.
Google Business Profile
A free listing that lets businesses manage how they appear on Google Search and Maps.
Why it matters when talking to AI: The single most important local SEO asset. AI agents can help optimize it but can't manage the verification.
Technical SEO
Improving how a website is crawled, indexed, and rendered by search engines.
Why it matters when talking to AI: Crawl errors, site speed, schema markup — the engineering side of SEO that AI tools can audit automatically.
Keyword Research
Identifying the words and phrases people use to search for products or services.
Why it matters when talking to AI: The foundation of SEO strategy. AI can accelerate keyword research but the strategic interpretation is human.
PPC (Pay-Per-Click)
Advertising model where you pay each time someone clicks your ad.
Why it matters when talking to AI: Google Ads, Meta Ads — PPC is the paid side of search. AI optimizes bids but budget decisions are yours.
CTR (Click-Through Rate)
The percentage of people who click on a link or ad after seeing it.
Why it matters when talking to AI: Low CTR = your headline or ad copy isn't working. AI can A/B test variations to improve it.
CPC (Cost Per Click)
How much you pay for each click in a paid advertising campaign.
Why it matters when talking to AI: Lower CPC = more clicks for the same budget. AI bid management can reduce CPC over time.
Conversion Rate
The percentage of visitors who complete a desired action — purchase, sign-up, call.
Why it matters when talking to AI: The metric that connects traffic to revenue. 1000 visitors at 2% conversion = 20 leads.
Bounce Rate
The percentage of visitors who leave after viewing only one page.
Why it matters when talking to AI: High bounce rate = something is wrong. Slow page, wrong content, or bad user experience.
A/B Testing
Comparing two versions of a page or ad to see which performs better.
Why it matters when talking to AI: Don't guess — test. AI can run and analyze A/B tests faster than manual analysis.
Landing Page
A web page designed specifically for visitors from a particular campaign or source.
Why it matters when talking to AI: Every ad should point to a landing page, not your homepage. AI can generate and optimize them.
Funnel
The series of steps a user takes from first contact to conversion.
Why it matters when talking to AI: Awareness → Interest → Decision → Action. Understanding funnels is understanding your client's business.
CAC (Customer Acquisition Cost)
The total cost to acquire one new customer, including all marketing and sales expenses.
Why it matters when talking to AI: If CAC is higher than customer value, you're losing money. AI can help identify the cheapest acquisition channels.
LTV (Lifetime Value)
The predicted total revenue from a customer over their entire relationship with the business.
Why it matters when talking to AI: LTV must be higher than CAC. This is the math that determines if a business model works.
ROAS (Return on Ad Spend)
Revenue generated per dollar spent on advertising.
Why it matters when talking to AI: A ROAS of 4:1 means every $1 in ads generates $4 in revenue. The metric ad clients care about most.
Schema Markup
Structured data you add to a website so search engines understand the content better.
Why it matters when talking to AI: Rich snippets, star ratings, FAQ dropdowns in search results — all powered by schema markup.
Domain Authority
A score predicting how well a website will rank, based on backlinks and other authority signals.
Why it matters when talking to AI: Higher DA = easier to rank. New sites start at 1. Building authority takes time and real links.
📊 Business & Operations
MECE
Mutually Exclusive, Collectively Exhaustive — a framework for organizing information without overlap or gaps.
Why it matters when talking to AI: The consulting gold standard. Every category is distinct (no overlap) and together they cover everything (no gaps). Use it when structuring any analysis.
SOP (Standard Operating Procedure)
A documented step-by-step process for completing a specific task consistently.
Why it matters when talking to AI: If you can't write the SOP, you can't delegate it — not to a human and not to an AI agent.
PRD (Product Requirements Document)
A document outlining what a product should do, its features, and success criteria.
Why it matters when talking to AI: Before you build anything — a website, a tool, a campaign — write the PRD. It's the brief that prevents scope creep.
KPI (Key Performance Indicator)
A measurable value that shows how effectively you're achieving a business objective.
Why it matters when talking to AI: Calls, leads, revenue, rankings — KPIs are what you report to clients. Everything else is vanity.
ROI (Return on Investment)
The profit or loss from an investment, expressed as a percentage of the cost.
Why it matters when talking to AI: The ultimate question every client asks: 'Is this worth the money?' ROI is the answer.
OKR (Objectives and Key Results)
A goal-setting framework that links big objectives to measurable results.
Why it matters when talking to AI: Objective: 'Become the top-ranked roofer in Tampa.' Key Result: 'Rank #1 for 5 target keywords by Q3.'
MVP (Minimum Viable Product)
The simplest version of a product that delivers value and lets you learn from real users.
Why it matters when talking to AI: Don't build the perfect website. Build the MVP, launch it, learn from data, then improve.
SOW (Statement of Work)
A document defining project deliverables, timelines, milestones, and responsibilities.
Why it matters when talking to AI: The contract between you and the client. If it's not in the SOW, it's not in scope.
Scope Creep
When project requirements expand beyond the original agreement without adjusting budget or timeline.
Why it matters when talking to AI: The #1 profitability killer for agencies. Every 'small addition' that isn't in the SOW erodes your margin.
Retainer
An ongoing monthly agreement where a client pays for guaranteed access to your services.
Why it matters when talking to AI: Predictable revenue. Retainers let you plan capacity and invest in quality instead of chasing new clients every month.
NDA (Non-Disclosure Agreement)
A legal contract preventing parties from sharing confidential information.
Why it matters when talking to AI: Sign it before the strategy call. Especially important when AI agents have access to client data.
CRM (Customer Relationship Management)
A system for tracking client interactions, deals, and communications.
Why it matters when talking to AI: HubSpot, Salesforce, Monday.com — your CRM is where client relationships live. AI can read from it but shouldn't replace it.
Stakeholder
Anyone who has interest in or influence over a project's outcome.
Why it matters when talking to AI: The client, their marketing director, the developer, the agency owner — all stakeholders with different priorities.
Agile
A project management approach that breaks work into small, iterative cycles.
Why it matters when talking to AI: Instead of one big launch, deliver in sprints. Review, adjust, repeat. How modern agencies should operate.
Sprint
A time-boxed period (typically 1-4 weeks) where a team completes a defined set of tasks.
Why it matters when talking to AI: Two-week sprints for client work. Clear goals, clear deliverables, clear review at the end.
☁️ Cloudflare & Edge Computing
Cloudflare Workers
Serverless functions that run at the network edge, close to users. No servers to manage.
Why it matters when talking to AI: Workers can run AI inference at the edge — lower latency, no origin server needed.
Edge Computing
Processing data close to where it's generated instead of in a centralized data center.
Why it matters when talking to AI: Running AI closer to users means faster responses. Cloudflare has 300+ edge locations.
Cloudflare Pages
A platform for deploying static websites and JAMstack apps directly from Git repos.
Why it matters when talking to AI: This website runs on Cloudflare Pages — deploy from GitHub, global CDN, zero config.
CDN (Content Delivery Network)
A network of servers distributed globally that cache and deliver content from the nearest location.
Why it matters when talking to AI: Your AI-powered site loads fast everywhere because the CDN serves it from the closest edge node.
Cloudflare Tunnel
A secure connection from your server to Cloudflare without opening inbound ports.
Why it matters when talking to AI: Expose your local AI agent to the internet securely — no open ports, no firewall rules.
DNS (Domain Name System)
Translates domain names (like truewebmaster.com) into IP addresses computers use.
Why it matters when talking to AI: Cloudflare DNS is often the first step — fast DNS means faster everything.
WAF (Web Application Firewall)
Filters and monitors HTTP traffic to protect web applications from attacks.
Why it matters when talking to AI: Protects your AI endpoints from injection attacks, rate abuse, and bot traffic.
SSL/TLS
Encryption protocols that secure data in transit. The padlock in your browser.
Why it matters when talking to AI: Cloudflare gives you free SSL — your AI-powered site is encrypted by default.
Reverse Proxy
A server that sits between users and your origin server, handling requests on its behalf.
Why it matters when talking to AI: Cloudflare acts as a reverse proxy — caching, security, and load balancing before traffic reaches your AI server.
KV (Key-Value Store)
A simple database that stores data as key-value pairs at the edge.
Why it matters when talking to AI: Store user sessions, config, or cached AI responses at the edge for instant access.
R2 (Object Storage)
Cloudflare's S3-compatible storage with zero egress fees.
Why it matters when talking to AI: Store AI model files, embeddings, or generated assets without paying to serve them.
Anycast
Network routing that directs requests to the nearest available server automatically.
Why it matters when talking to AI: Type a URL in Tokyo, your request goes to the Tokyo edge server. Type it in Miami, it goes to Miami. Automatic.
Cold Start
The delay when a serverless function runs for the first time — it needs to spin up before executing.
Why it matters when talking to AI: First request is slow, subsequent requests are fast. Workers and edge functions minimize this.
Edge Functions
Serverless functions deployed to CDN edge locations — run your code close to users worldwide.
Why it matters when talking to AI: Your AI preprocessing runs in 300+ locations simultaneously. No origin server round-trip needed.
No glossary terms matched your search. Try a broader word like API, token, memory, or server.