Building sites and apps with AI

In short

Today you can build a site or small app by describing it in words. For prototypes it’s revolutionary; for production it takes method.
The most complete platforms are Claude Code (the best) and Codex: very powerful models included in reasonably priced subscriptions. Then Cursor, Google Antigravity, Lovable and Replit.
They are now so polished that you can use them even if you can’t code at all: just describe clearly what you want.
The secret is a clear written description before you start, and moving in small steps.
Golden rule: never put secret keys in code that runs in the browser, and have security and data reviewed before going live.

From “vibe coding” to agentic engineering

In 2025 Andrej Karpathy coined vibe coding: building software by describing what you want and accepting the code the AI writes, without even reading it. It works surprisingly well for prototypes, personal pages, internal tools.

In 2026 the conversation moved on: coding agents work for hours on real projects, write tests, run them, fix errors. Karpathy himself now talks about agentic engineering: you no longer write code line by line, you design, instruct and verify. The craft changes; it doesn’t disappear.

According to Arena’s WebDev ranking (October 2026), where people vote for the better site generated from the same prompt, the strongest models are Claude Opus 5.5 (1,811 points), GPT-6 Astra (1,780), Claude Sonnet 5.5 (1,774) and GPT-6.1 Sol (1,760).

Which tool

1. Claude Code, my first choice. In my experience it is the most complete platform: it builds whole sites and apps, works on your files, tests the result in the browser, fixes errors, and beyond code it handles automation, documents and analysis. It uses Claude models, the best at coding (Opus 5.5 leads the WebDev Arena), and it comes with the Claude subscription: for about €20 a month you get models that would cost far more via API.

2. Codex, the runner-up. OpenAI’s equivalent, included in ChatGPT plans: very capable, it can work on several tasks in parallel in the cloud, handy if you already use ChatGPT.

3. Cursor, if you like seeing the code. A code editor with built-in agents: it writes, edits and checks the project, and lets you choose among the best models (Claude included). It’s built for people who want the files in front of them; it’s now also included in Grok’s X Subscription.

The others, briefly:

Tool	For	Good to know
Google Antigravity	People in Google’s ecosystem	Agentic environment with Gemini 4 Argon; limits tied to your Google AI plan
Lovable	A live prototype in an hour	Very simple; deep customisation is harder
Replit	An app with a database, all in the browser	Hosting and database included, billed separately

Claude Code and Codex are a little more complex than Lovable and Replit, because they do much more: at first a ready-made environment in the browser may feel easier. But Claude Code, Codex and Antigravity are now so well optimised, and the models so good, that you can use them comfortably without knowing how to code: describe what you want in plain words, the agent proposes a plan, builds, tests and shows you the result. It takes method, not coding skills.

This site is an example

This site was built with a coding agent (Claude Code): structure, design, interactive components, chatbot. But every choice went through a written description, checks in the browser and precise security rules, such as the chatbot’s key staying on the server.

The five-step method

1. Write what you want before opening the tool

A one-page document is worth hours of fixes. Who will use it, what they must be able to do, which pages or screens, what data it stores, how it should look, what it must NOT do.

2. Hand over the whole result, then refine

With today’s agents you don’t need to go one feature at a time: describe the complete app, with material and examples, and let the agent propose a plan and carry it all out, testing in the browser on its own. Your job is to read the plan before it starts and correct what doesn’t add up. The place to stop is at the end: test everything like a real user before publishing.

3. Test like a real user

Click everything, on desktop and phone. Enter wrong data, empty fields, very long text. Describe problems to the agent precisely: what you did, what you expected, what happened, the error message.

4. Save versions

Use Git (or the tool’s version history). When the AI breaks something that worked, and it will, you roll back in a second.

5. Publish calmly

Services like Vercel, Netlify or Cloudflare publish a site in minutes, often free for small projects. Before sharing it, go through the list below.

Security: the checklist not to skip

Secret keys only on the server. An API key written into browser code can be copied by anyone opening the page, and someone will use your credit. Keys go into the server’s environment variables.
Spending limits. Set a monthly cap in the AI provider’s dashboard.
Access and permissions. Who can see what? A user must not be able to read another user’s data by changing a number in the URL.
Personal data. Collect only what’s needed, write a privacy notice, pick providers with suitable servers and contracts.
Dependencies. AI sometimes installs packages that don’t exist or are abandoned. Check that the ones used are well-known and maintained.
Review. For anything handling money, health data or customer data, have a developer review the code before production.

When you need a developer

Payments, health or financial data, many concurrent users, integrations with company systems: here AI speeds work up enormously, but you need someone who can assess it. The cost of a mistake far exceeds the savings.

What it costs

Prototype: Claude Code comes with Claude Pro (about €20 a month), Codex with ChatGPT plans; Lovable has a free plan with limited credits.
Regular use: ~€20/month plans are enough for small projects; for intensive all-day sessions you need higher plans or API credits.
Publishing: for a site like this, free or a few euros a month, plus any AI APIs used (this site’s chatbot costs a few cents per question).

In practice

Try this week: take an Excel sheet you use often (a quote calculator, a recurring calculation) and ask Claude Code to turn it into a small web app. It’s the perfect exercise: real problem, simple data, verifiable result.

Sources

Informational content, current as of the date shown. Prices, models and rules change often: always check official sources. Written with the support of AI tools and reviewed by the author.