Plain English · Glossary

Tech, explained simply.

59 terms that get nodded at and never explained — API, CRUD, frontend, backend, cloud, LLM and the rest. We wrote this so clients, students and the simply curious can follow any tech conversation. No gatekeeping.

Foundations

Frontend
Everything you see and tap — the screens, buttons and layouts. It runs on your phone or in your browser.
Backend
The part you never see: the servers, logic and databases that do the work after you tap. The kitchen behind the restaurant.
Full-stack
Work that spans both frontend and backend — and an engineer comfortable across the whole thing.
Server
A computer, always on somewhere, whose job is to answer requests from apps and browsers.
Database
Where an app stores its information so it's still there next time — users, orders, messages, all of it.
Cloud
Renting computers and storage from a big provider instead of owning the hardware. 'The cloud' is just someone else's data center.

How things talk

API
A defined way for two pieces of software to talk. Like a waiter: you ask for something specific, it brings it back — you never see the kitchen.
Endpoint
One specific address on an API you can call — e.g. 'give me this user's orders.'
CRUD
Create, Read, Update, Delete — the four basic things almost every app does to its data. Add a note, view it, edit it, delete it.
JSON
A simple, readable text format apps use to send data to each other. Most APIs speak it.
Webhook
A reverse API call — instead of you asking, a service pings you the moment something happens ('a payment just arrived').
HTTPS / SSL
The padlock in your browser. It encrypts the connection so no one in between can read what's sent.
REST
The most common style of API — predictable web addresses for your data, using plain web verbs to get, add, change or remove it.
Request / Response
The basic exchange of the web: your app sends a request, the server sends a response back. Every tap is a tiny conversation.

Building blocks

Framework
A pre-built foundation that handles the repetitive parts so engineers build features, not plumbing. Next.js and Flutter are frameworks.
Library
A reusable bundle of code you drop in to do one job — a date picker, a chart — without writing it yourself.
SDK
Software Development Kit — a toolbox a company gives you to plug their service (payments, maps) into your app.
Open source
Code anyone can read, use and improve for free. Much of the internet runs on it.
Repository (repo)
The folder where a project's code lives and its full history is tracked. Usually on GitHub.
Git
The system that tracks every change to code, so teams can work together and undo mistakes safely.
Source code
The human-readable instructions engineers write. It gets run or compiled to become the working app.
Version control
Keeping a full, rewindable history of every change to the code, so nothing is ever truly lost. Git is the standard.
Pull request (PR)
A proposed change to the code, opened for teammates to review before it joins the main project.
Environment
A separate copy of the app for a purpose — 'development' to build in, 'staging' to test, 'production' for real users.

Kinds of apps

Native app
An app built specifically for iOS or Android, installed from the App Store. Fastest, deepest access to the phone.
Cross-platform
Writing the app once and running it on both iOS and Android — saving time and budget.
Web app
An app that runs in the browser, nothing to install. Gmail is a web app.
PWA
Progressive Web App — a website that can be installed and works offline, blurring the line with native apps.
UI
User Interface — the visual design: what the product looks like.
UX
User Experience — how the product feels to use: is it clear, fast, frustrating, delightful?

Shipping & ops

Deployment
Putting code live so real people can use it. 'We deployed the fix' = it's now in your hands.
Hosting
Where an app actually runs and is served from — the always-on home for its code and files.
CI/CD
Automation that tests code and ships it safely every time it changes — so releases are routine, not scary.
Cache
A temporary copy kept close by so things load instantly instead of being fetched again. Why a page is faster the second time.
Latency
The delay between asking for something and getting it. Low latency feels instant; high latency feels laggy.
Bug
A mistake in the code that makes the software behave wrong. Fixing them is most of engineering.
Container
A lightweight box that packages an app with everything it needs to run, so it behaves the same on any machine. Docker made them mainstream.
Serverless
Running code without managing servers yourself — the cloud spins it up on demand and you pay only for what runs.
Monitoring
Watching a live app's health — errors, speed, traffic — so problems are caught before users complain.
Scalability
How well an app copes as usage grows from a hundred users to a million without falling over.

Data & AI

Algorithm
A precise set of steps to solve a problem — like a recipe a computer follows exactly.
Machine learning
Software that finds patterns in examples instead of being told every rule — the basis of modern AI.
LLM
Large Language Model — the kind of AI behind chat assistants. It predicts text well enough to answer, summarize and write code.
Token
The small chunks of text an AI reads and writes in. Roughly ¾ of a word each — and what AI usage is billed by.
Authentication
Proving who you are — logging in. (Authorization is the next step: what you're allowed to do.)
RTL
Right-to-Left — designing interfaces to read correctly in Arabic and other RTL languages, not just mirrored English.
Prompt
The instruction you give an AI model. Clear, specific prompts get better results — writing them well is its own craft.
RAG
Retrieval-Augmented Generation — giving an AI your own documents to answer from, so it cites real facts instead of guessing.
Inference
Actually running a trained AI model to get an answer. Each inference costs compute — and money.
Hallucination
When an AI states something false with confidence. Why human review (and RAG) matter for anything that counts.
Vector database
A database that stores meaning as numbers, so an AI can find 'similar' things — the memory behind RAG and smart search.
Fine-tuning
Further-training a general AI model on your own data so it specializes in your domain, format or style.

Quality & reach

Responsive design
Layouts that adapt to any screen — phone, tablet, desktop — instead of breaking on the small ones.
Accessibility (a11y)
Designing so everyone can use the product — keyboard, screen reader, low vision, reduced motion. A requirement, not an extra.
SEO
Search Engine Optimization — building a site so Google understands and ranks it, bringing visitors without paid ads.
Performance budget
A hard limit on how heavy a page is allowed to be, so it stays fast on cheap phones and slow networks.

Security & data

Authorization
What you're allowed to do once logged in — an admin sees more than a guest. The step after authentication.
Encryption
Scrambling data so only the right key can read it — protecting it both in transit and at rest.
GDPR
The EU's data-protection law: rules for handling personal data — consent, access, deletion — with real penalties for getting it wrong.
The process

When each piece shows up.

A product gets built in roughly this order. Tap any term to jump to its plain-English definition.

  1. 01

    Design

    Decide what to build and how it looks and feels — before a line of code.

  2. 02

    Set up

    Create the project, its home, and a way to track every change.

  3. 03

    Build the frontend

    The part users actually see and tap, in the browser or app.

  4. 04

    Build the backend

    The engine behind the screen — logic, data and the ways they talk.

  5. 05

    Add intelligence

    Optional: weave in AI features, grounded in real data.

  6. 06

    Review & test

    Catch problems early, while they're cheap to fix.

  7. 07

    Ship it

    Make it live for real people — safely and repeatably.

  8. 08

    Run & grow

    Keep it fast and healthy as more people show up.

Still stuck on a word?

We'd rather answer the question than have you nod along. Ask us anything — no such thing as a basic one.