The two guys wrote an AI application overnight, earning 20,000 a month? ? (A development tutorial is attached at the end of the article)

Developing a tool that can talk to AI to generate and edit mind maps sounds like something that can only be accomplished by a group of professional AI background teams spending a lot of time and energy training models and polishing applications.

However, two college students completed it overnight. Just like alchemists transforming vulgar materials into gold, they transformed the code into a magical tool. Let's uncover the secret behind this magical tool together.

use

Case 1: Travel strategy.

The most troublesome part of traveling is to make a strategy, which places to visit, how to arrange the itinerary, etc. Then as long as we use this magical tool, things will become easier immediately.

Input: Yunnan Travel Guide

How about it? Do you want to go with a backpack right away?

Case 2: Design Creativity

A tool that cannot solve productivity is definitely not a good tool. Let's see how it can help with interior design.

Enter: Cyberpunk-Inspired Interior Design Ideas

Here comes the inspiration!

to develop

According to the author, they used Laf as the development platform, and then used ChatGPT3.5 for natural language processing and AI interaction.

That is to say, use Laf to interface with ChatGPT and add Yidiandian front-end details. After the development is completed, it will be hosted on Laf, and such an application will be launched!

Is it fast enough to launch an application overnight? It should be fast enough, but it's the limit of two college students, not Laf's limit.

So how fast is Laf? The following is the whole process of writing a ChatGPT interface.

1. Open the browser and log in to laf.dev

2. Create a new cloud function, select ChatGPT as a template

3. Change apiKey to your own

4. Click Publish in the upper right corner

ok An interface is already online, the front end directly requests this interface through http, and when the text is passed in, it will get a reply from ChatGPT, and it supports streaming return (typewriter effect).

With the increase of function templates and the subsequent launch of the function market, basically 80% of the business logic can be completed by clicking and changing, and bid farewell to reinventing the wheel.

online

After the code is written, it will go online. It stands to reason that you need to buy a server, configure a proxy, etc., but these are not fast enough. Using this method, two guys can’t go online overnight.

Then the fastest way is to create a Bucket on Laf, throw the packaged project directly into it, and click hosting to go online.

I would like to call it: second-level development, second-level online, it is so fast!

Summarize

As the saying goes: the only thing that can't be broken in martial arts in the world is speed. In this era of unpredictable changes, you can only be one step ahead if you are fast enough. If you have a good idea, you must realize it quickly.

Through Laf, developers do not need to pay attention to the underlying server, agent and other operation and maintenance issues, but can focus on application development and the realization of ideas, and can quickly transform ideas into products, quickly respond to market changes, and thus gain a competitive advantage.

Finally, let us shout out the resounding slogan: life is short, you need laf :)

Laf: https://github.com/labring/laf ChatMind: https://www.chatmind.tech Launch your own ChatGPT tutorial in three minutes: https://forum.laf.run/d/100

Guess you like

Origin blog.csdn.net/alex_yangchuansheng/article/details/130520456