What is AI?
A plain-English introduction to artificial intelligence for developers.
Artificial Intelligence (AI) is software that performs tasks we normally associate with human thinking: understanding language, recognising images, making predictions, and generating new content.
This guide is about a specific kind of AI
Modern AI apps are mostly built on Large Language Models (LLMs) — systems that read and write text. As a developer you rarely train these models yourself. Instead you call them over an API, send instructions, and use their output in your app.
What you will learn
- How LLMs work at a practical level.
- How to call an LLM API and shape its output with prompts.
- Embeddings, semantic search, and Retrieval-Augmented Generation (RAG).
- Tool use and building AI agents.
- How to ship AI features safely and cheaply.
Example
# The whole guide in one idea: send text to a model, get text back.
curl https://api.example-llm.com/v1/chat \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "llm-medium",
"messages": [{"role": "user", "content": "Explain AI in one sentence."}]
}'When to use it
- A startup uses AI to automatically classify incoming support tickets so the right team handles each one instantly.
- An e-commerce site uses AI-powered recommendations to surface products each visitor is likely to buy.
- A hospital uses AI to flag abnormal lab results in patient records, alerting doctors before they even open the file.
More examples
AI classifies text input
Sends a single message to a chat model and gets a one-word classification back.
curl https://api.openai.com/v1/chat/completions \
-H 'Authorization: Bearer $OPENAI_API_KEY' \
-H 'Content-Type: application/json' \
-d '{
"model": "gpt-4o-mini",
"messages": [{"role":"user","content":"Is this a complaint or a question? One word.\nI never received my order."}]
}'AI summarises a document
Uses a system prompt to constrain the task, then passes the document as the user message.
curl https://api.openai.com/v1/chat/completions \
-H 'Authorization: Bearer $OPENAI_API_KEY' \
-H 'Content-Type: application/json' \
-d '{
"model": "gpt-4o-mini",
"messages": [
{"role":"system","content":"Summarise the text in one sentence."},
{"role":"user","content":"Artificial intelligence is the simulation of human intelligence..."}
]
}'Python SDK minimal AI request
Shows the minimal Python SDK call pattern: import, create client, send message, read reply.
from openai import OpenAI
client = OpenAI() # reads OPENAI_API_KEY from env
response = client.chat.completions.create(
model="gpt-4o-mini",
messages=[{"role": "user", "content": "Name three uses of AI in healthcare."}]
)
print(response.choices[0].message.content)
Discussion