The AI Glossary
Plain-English definitions of the concepts behind AI-native organizations, infrastructure, hiring, search and strategy, for executives, not engineers.
AI Infrastructure
Large Language Model (LLM)
A model trained on vast text to predict and generate language, the engine behind tools like ChatGPT and Claude. It powers chat, drafting, extraction and reasoning over natural language.
Retrieval-Augmented Generation (RAG)
A pattern that grounds an LLM in your own, current data by retrieving relevant documents at query time, reducing hallucination and keeping answers up to date without retraining.
Vector Database
A database that stores embeddings (numeric representations of meaning) and finds the most similar items fast, the backbone of semantic search and RAG.
Embedding
A numeric vector that captures the meaning of text (or images) so machines can compare similarity. Embeddings power search, recommendations and RAG retrieval.
Fine-tuning
Further training a model on your examples to change its behavior, style or format. Best for consistent behavior, not for injecting fresh facts (use RAG for that).
Prompt Engineering
The practice of structuring instructions and context so a model reliably produces the output you need. It is craft, not magic, and increasingly systematized.
AI Agent
An LLM-driven system that plans and takes multi-step actions using tools (APIs, search, code). Powerful for bounded, supervised workflows; risky when unsupervised in high-stakes tasks.
Model Context Protocol (MCP)
An open standard for connecting AI models to tools and data sources through a common interface, so agents can use your systems without bespoke integrations for each one.
MLOps
The practices and tooling that make machine-learning models reliable in production, deployment, monitoring, versioning, retraining and cost control.
Inference
Running a trained model to produce an output. Inference cost and latency, not training, dominate the economics of most production AI features.
Token
The unit of text an LLM processes, roughly a word-piece. Usage and pricing are measured in tokens, so prompt and output length directly drive cost.
Context Window
The maximum amount of text (in tokens) a model can consider at once. Larger windows allow more context but cost more and can dilute focus.
Hallucination
When a model produces confident but false or unsupported output. Mitigated with retrieval (RAG), grounding, citations and human review in sensitive contexts.
Guardrails
Controls that constrain what an AI system can say or do, input/output filtering, policy checks, least-privilege tool access, so it behaves safely in production.
Foundation Model
A large, general-purpose model trained on broad data that can be adapted to many tasks, the base you build on rather than train from scratch.
Multimodal AI
Models that work across more than one data type, text, images, audio, video, enabling use cases like document understanding and image-based support.
Human-in-the-Loop
A design where a person reviews or approves AI output before it acts. Essential for high-stakes workflows and a safe way to widen automation over time.
AI Observability
Monitoring the behavior, quality, cost and safety of AI systems in production, logging prompts, outputs, latency and failures, so you can debug and improve them.
Evaluations (Evals)
Systematic tests that measure an AI system's quality on representative tasks. Good evals turn 'it feels better' into evidence and prevent silent regressions.
Agentic Workflow
A process where an AI agent chains multiple steps and tools toward a goal. Works best when bounded, observable and reversible, with a human on high-stakes actions.
AI Search (GEO)
Generative Engine Optimization (GEO)
Optimizing to be cited and well-represented inside AI-generated answers (ChatGPT, Perplexity, Gemini, AI Overviews), across the diverse sources those engines synthesize.
Answer Engine Optimization (AEO)
Optimizing content to be the extracted answer to a specific question, in snippets, voice and AI results, through clear structure, direct answers and schema.
Search Engine Optimization (SEO)
Earning visibility on traditional search results through relevant content, technical health and authority. Still the foundation that feeds the indexes AI engines draw from.
Share of Voice (AI Search)
The share of relevant AI answers that cite your brand versus competitors, the closest thing to a ranking in AI search, and a core GEO KPI.
AI Overviews
Google's AI-generated summaries shown above traditional results. Being cited there is prime visibility, and doesn't simply follow from ranking first.
Schema Markup
Structured data (schema.org, usually JSON-LD) that tells machines what your content means, helping search and AI engines extract accurate answers and attribute them to you.
AI Hiring
Forward-Deployed Engineer
A senior engineer who embeds in your team and ships to production, transferring patterns your team can own, rather than delivering from arm's length.
Fractional Executive
A senior leader (CTO, CPO, CISO, CMO) on a part-time retainer who owns direction and first hires, without the cost and commitment of a full-time exec.
Staff Augmentation
Extending your team with external specialists who work under your direction, keeping ownership and knowledge in-house, versus outsourcing a whole deliverable.
ML Engineer
An engineer who builds and improves machine-learning models, data, features, evaluation, accuracy, and ships them toward production.
AI Product Manager
A product manager fluent in data, model evaluation and AI's limits, who turns model capability into usable, valuable product, and knows when not to use AI.
Time-to-Value
How quickly a hire or initiative delivers real, measurable value. Embedded and fractional models optimize for days-to-weeks rather than quarters.
AI Strategy
AI-Native Organization
A company structured to exploit AI in how it hires, builds and operates, sequencing capability before tooling, not one that merely buys AI products.
Return on Investment (ROI)
The measurable value an AI initiative returns versus its cost. Model it per use case, impact minus build-and-run cost, before committing budget.
Payback Period
How long until an initiative's cumulative value covers its cost. For operational AI, a payback inside ~12 months is a strong signal to proceed.
Total Cost of Ownership (TCO)
The full cost of an AI capability over time, build plus run, compute, monitoring, maintenance and iteration, not just the initial project or license.
Build vs Buy vs Embed
The decision, per capability, to build in-house (differentiators), buy a vendor (commodity) or embed specialists (core but you lack capacity now).
EU AI Act
The EU's risk-based AI regulation. Obligations, transparency, documentation, human oversight, scale with the risk category of the use case. (Not legal advice.)
Proof of Concept (PoC)
A small build to test whether an AI approach works before committing. Design it with production concerns in mind, or it becomes a demo that never ships.
Churn
The rate at which customers stop paying. AI can predict at-risk accounts early, but retained revenue comes from the human intervention that follows.
Lead Scoring
Ranking prospects by likelihood to convert so sales focuses its time. AI scoring works when it is explainable and tied to real outcomes.
The Weekly Brief
Intelligence for building AI-native organizations.
One email a week: the sharpest thinking on AI hiring, infrastructure, teams and strategy, for the people building the future of work.
Join operators, founders and CTOs. No spam, unsubscribe anytime.