Future Skills · Glossary

AI & Tech Glossary

63 terms explained in plain English, each with an example and a link to the full lesson.

Search terms

All Artificial Intelligence Prompt Engineering Python AI Agents Data Analytics Machine Learning Cloud Computing Cybersecurity

A

AI AgentAn AI system that can plan steps and use tools — search, code, apps — to complete a task with less step-by-step instruction.Artificial Intelligence Artificial Intelligence (AI)Software that learns patterns from data and uses them to make predictions, decisions or new content.Artificial Intelligence Availability ZoneAn isolated data centre location within a cloud region, with independent power, cooling and networking.Cloud Computing

B

Bias (in AI)Systematic unfairness in AI output, usually inherited from imbalanced or historical training data.Artificial Intelligence

C

Chain-of-Thought PromptingAsking a model to reason through intermediate steps before giving its final answer.Prompt Engineering CIA TriadThe three core goals of information security: confidentiality, integrity and availability.Cybersecurity ContainerA lightweight package containing an application and everything it needs to run, sharing the host operating system’s kernel.Cloud Computing Context WindowThe maximum amount of text a language model can consider at one time.Artificial Intelligence Cross-Site Scripting (XSS)An attack where untrusted input is shown on a page without encoding, so an attacker’s script runs in other users’ browsers.Cybersecurity Cross-ValidationEvaluating a model by training and validating it several times on different splits of the data and averaging the results.Machine Learning

D

Data CleaningFixing or removing incorrect, duplicate, inconsistent and missing data so analysis can be trusted.Data Analytics Data LeakageWhen information unavailable at prediction time sneaks into training, making a model look better than it really is.Machine Learning DataFrame (pandas)A table of rows and columns in Python’s pandas library, used to clean, analyse and reshape data.Data Analytics Deep LearningMachine learning that uses neural networks with many layers, especially good for images, audio and language.Artificial Intelligence DelimitersMarkers like triple quotes or <tags> that separate instructions from the text you want processed.Prompt Engineering DictionaryA Python collection that stores key-value pairs for fast lookup by key.Python

E

EmbeddingA list of numbers that represents the meaning of a piece of text, so similar meanings are close together.AI Agents ExceptionAn error that happens while a program is running, which can be handled with try / except.Python

F

Feature EngineeringCreating, transforming and selecting input variables so a model can learn patterns more easily.Machine Learning Few-Shot PromptingGiving a model a few input-output examples before the real input so it copies the pattern.Prompt Engineering Fine-TuningFurther training a pre-trained model on a smaller, specific dataset to adapt it to a task or style.Artificial Intelligence FunctionA named, reusable block of code that takes inputs (parameters) and can return a result.Python

G

Generative AIAI that creates new content — text, images, code or audio — instead of only labelling or predicting.Artificial Intelligence

H

HallucinationWhen an AI model confidently states something false or made up.Artificial Intelligence Human-in-the-LoopA design where a person reviews or approves an AI system’s action before it takes effect.AI Agents

I

IaaS, PaaS and SaaSThe three cloud service models: rent infrastructure, use a managed platform for your code, or use complete software.Cloud Computing IAM (Identity and Access Management)The cloud service that controls who can do which actions on which resources.Cloud Computing

K

KPI (Key Performance Indicator)A number that shows how well a business or team is doing against an important goal.Data Analytics KubernetesAn open-source system for running, scaling and updating containers across a cluster of machines.Cloud Computing

L

Large Language Model (LLM)A deep learning model trained on huge amounts of text to predict the next token, which lets it understand and generate language.Artificial Intelligence Linear RegressionA model that predicts a number as a weighted sum of input features — a best-fit straight line or plane.Machine Learning List ComprehensionA one-line way to build a list from another sequence, optionally with a condition.Python

M

Machine Learning (ML)A way of building AI where the system learns rules from examples instead of being programmed with them.Artificial Intelligence Model Context Protocol (MCP)An open standard for connecting AI applications to tools and data sources through reusable servers.AI Agents Multi-Factor Authentication (MFA)Logging in with two or more different kinds of proof: something you know, have or are.Cybersecurity

N

Neural NetworkA model made of layers of connected units that transform input into output, loosely inspired by brain cells.Artificial Intelligence

O

OverfittingWhen a model memorises its training data so closely that it performs poorly on new data.Artificial Intelligence

P

Password HashingStoring a one-way, salted, deliberately slow transformation of a password instead of the password itself.Cybersecurity PhishingA fraudulent message pretending to come from a trusted source to steal credentials, money or access.Cybersecurity Pivot TableA spreadsheet tool that summarises data — sums, counts, averages — grouped by categories, by dragging fields.Data Analytics Precision and RecallTwo classification metrics: precision is how many flagged items were correct; recall is how many actual positives were caught.Machine Learning PromptThe input you give an AI model — instructions, context, examples and data.Prompt Engineering Prompt EngineeringThe skill of writing and improving prompts so AI tools give accurate, useful output.Prompt Engineering Prompt InjectionAn attack where text inside the input tries to override the AI’s original instructions.Prompt Engineering

R

RAG (Retrieval-Augmented Generation)A technique where relevant documents are retrieved and given to the model so it answers from them instead of memory.Artificial Intelligence Random ForestAn ensemble model that combines many decision trees trained on random samples of data and features.Machine Learning ReAct PatternAn agent pattern that alternates reasoning about the next step with acting through a tool, then observing the result.AI Agents

S

ServerlessA cloud model where the provider runs your code on demand and bills per request and run time, with no servers for you to manage.Cloud Computing Shared Responsibility ModelThe split of security duties between a cloud provider (security of the cloud) and the customer (security in the cloud).Cloud Computing SQL InjectionAn attack where input changes the meaning of a database query because it was inserted directly into the SQL text.Cybersecurity SQL JOINA SQL operation that combines rows from two tables based on a related column.Data Analytics Supervised LearningTraining a model on examples that include the correct answer (labels).Artificial Intelligence System PromptBackground instructions that set an AI assistant’s role, rules and tone for a whole conversation.Prompt Engineering

T

TemperatureA setting that controls how random or predictable a language model’s output is.Artificial Intelligence TokenA small chunk of text — a word, part of a word or punctuation — that a language model processes as one unit.Artificial Intelligence Tool Calling (Function Calling)A model capability where the AI returns a structured request to run a function with specific arguments, instead of plain text.AI Agents Train-Test SplitDividing data into a part for training a model and a held-out part for checking how it performs on unseen data.Machine Learning

U

Unsupervised LearningTraining a model on data without labels so it finds structure on its own.Artificial Intelligence

V

VariableA name that refers to a value stored in memory.Python Vector DatabaseA database designed to store embeddings and quickly find the most similar ones to a query.AI Agents

W

Window FunctionA SQL function that calculates across related rows without collapsing them into one row.Data Analytics

Z

Zero TrustA security model that never grants trust based on network location and verifies every user, device and request.Cybersecurity Zero-Shot PromptingAsking a model to do a task without giving any examples.Prompt Engineering