Building apps
What is low-code?
Low-code describes platforms that build most of an application visually while allowing custom code where the visual tools fall short, aimed at teams that have some development capability but want to move faster.
New to Bolt? Sign up free.
Technical definition
Low-code platforms provide visual application assembly with defined extension points where custom code can be written: expressions, event handlers, API calls, and custom components. The platform manages the application shell, the data layer, and deployment while permitting code where configuration is insufficient.
The distinction from no-code is the intended user. No-code targets a business user with no programming capability. Low-code targets a developer or technical operator who wants the platform to handle the routine parts.
In plain terms
Low-code is visual building with an escape hatch. Configure most of it, write code for the bits that need it.
Which means it assumes someone on your side can write that code.
Why it matters
The name misleads people. Low-code sounds like a gentler version of coding, so non-technical buyers pick it and then find that the parts they most need are behind the escape hatch they cannot use.
The honest positioning of the three categories:
- No-code works with no technical skill, and stops at the platform's ceiling.
- Low-code goes further, and needs someone who can write the extension code.
- AI builders need no technical skill and have no component ceiling, because they generate code rather than configuration.
What all of the older categories share, and what is worth checking before you commit, is portability. A low-code application lives inside its platform. The custom code you wrote is embedded in that platform's structure, and it does not come out as a standalone project.
