DEV Community: Chiheb Nabil

Why I built Aipify. co - A Platform for building AI Powred Apis

Chiheb Nabil — Fri, 01 Mar 2024 11:04:49 +0000

One of the biggest challenges the developer may face is structuring ChatGPT responses when building production-ready apps. You want your responses to be consistent since the app's frontend is expecting a specific field. With OpenAI, you have two main ways to achieve that:

Learning prompt engineering techniques to achieve the desired structure.
Using the function calling feature.

The problem is that if you want to change anything, you have to alter that hardcoded prompt and push it again. Additionally, if you want to cache a specific response to avoid additional costs, you have to implement that on your backend.

I learned this the hard way when building the first version of Compass.

Step by step, I found myself building a whole SaaS service that I'm currently using for all my ChatGPT projects and ideas.

This is how the Aipify idea was born. After over one month of building and tweaking, this is the version of Aipify, and guess what? It has a monthly free quota.

You can experiment with GPT-4 for free, and I'm eager to hear your feedback 😄

Why I Built the SES Easy Mailer Node Module

Chiheb Nabil — Thu, 01 Feb 2024 13:17:06 +0000

As a software developer, I often find myself in situations where I need to send emails from my applications.

Whether it's for sending notifications, password resets, or promotional content, email is a crucial part of many web applications.

Amazon's Simple Email Service (SES) is a reliable and scalable solution for sending emails, but it can be a bit complex to use directly. That's why I decided to build the SES Easy Mailer Node module.

Making Email Sending Easier with SES

The SES Easy Mailer module simplifies the process of sending emails using Amazon SES.

It provides a simple and easy-to-use interface for sending emails, abstracting away the complexities of the underlying SES APIs.

This makes it a great choice for developers who want to leverage the power of Amazon SES without having to deal with its intricacies.

Here's a simple example of how to send an email with SES Easy

const { createTransporter, sendMail } = require('ses-easy-mailer');
const { SESClient } = require('@aws-sdk/client-ses');

const client = new SESClient(
    region: "us-east-1",
    credentials: {
        accessKeyId: "",
        secretAccessKey: "",
    }
);
const transporter = createTransporter(client);

Using Local Html Template

const from = 'sender@example.com';
const to = 'recipient@example.com';
const subject = 'Hello, world!';
const templateType = 'file';
const templatePath = './template.html';
const templateData = { name: 'John Doe' };
const attachments = [];

await sendMail(
    transporter,
    from,
    subject,
    templateType,
    templatePath,
    templateData,
    attachments,
    to
)

Using SES Template

On the other hand, if you prefer to use SES's built-in template system, SES Easy Mailer has got you covered.
You can create and manage your templates directly in SES and simply reference them by name when sending emails. This is a great option for developers who want to manage their templates separately from their code.

Here's an example of how to send an email using an SES template:

const templateType = 'ses';
const templateName = 'se-template-name'; // The name of the template you created in SES

await sendMail(
    transporter,
    from,
    subject,
    templateType,
    templateName,
    templateData,
    attachments,
    to
)

Simplifying Attachments and Templating

Adding attachments to your emails is as simple as passing an array of objects, each representing an attachment.
The module takes care of the rest, ensuring that your attachments are correctly included in the sent email.

Here's an example of how to send an email with attachments :

let buffer = Buffer.from("hello world!", "utf-8");

let attachments = [
    {
        filename: "test.txt",
        content: buffer,
        encoding: "base64",
    }
]

Conclusion

Building the SES Easy Mailer Node module has been a fun and rewarding experience. It's always great to build something that not only solves your own problems but can also be useful to other developers. If you're looking for a simple and easy-to-use solution for sending emails with Amazon SES, I encourage you to give SES Easy Mailer a try !

OpenAI API Mock for Devs on a Budget

Chiheb Nabil — Mon, 29 Jan 2024 15:27:54 +0000

Hey, fellow devs! Ever been working on a project, eager to test your OpenAI integration, only to be held back by the dreaded API call limits? Well, fear not! I've got a nifty solution for you, and it involves a touch of magic and a dash of mock-tastic sorcery.

🧙‍♂️ Introducing OpenAI API Mock

Say hello to openai-api-mock, a Node.js module that's here to turn your development environment into a playground for all your OpenAI API dreams. No more worrying about unnecessary API charges during testing – it's time to unleash the power of mock responses!

🌟 Why Did I Create This Module?

Picture this: you're knee-deep in developing a project, excited to test your OpenAI integration. However, you hit a roadblock – the dreaded API call limits.

Here's the twist – I needed a way to mock API responses, especially when dealing with function-calling structured responses & API costs can spiral during development and testing phases, and I yearned for a tool that could empower developers to experiment freely with OpenAI without the fear of skyrocketing bills. Thus, openai-api-mock was born,

🎩 How to Work Your Mock Magic

Step 1: Install the Wizardry

Start by installing this enchanting module using npm. Open your terminal and type:

npm install -D openai-api-mock

Step 2: Summon the Mock Response

In your JavaScript file, conjure the mock response with a simple incantation:

const { mockOpenAIResponse } = require('openai-api-mock');

mockOpenAIResponse();

Now, every time your code attempts to make an OpenAI API call, it will receive a magical mock response. No API call limits, no charges – just pure wizardry at play!

🧐 What's Under the Hood?

This magical module depends on the powers of nock for intercepting HTTP calls and @faker-js/faker for generating fake data. It's a dynamic duo ensuring your OpenAI API calls are met with whimsical responses.

🚀 Conclusion
With the power of mock responses in your hands, you're ready to take your OpenAI journey to new heights without worrying about the cost. openai-api-mock is here to make your development experience enchanting and budget-friendly.

Github Repo