At SOMOS.tech, our mission has always been to break down barriers and ensure that everyone, regardless of their background, can build a fulfilling career in tech. It's with great pride that we introduce the latest milestone in our journey: the Starting in Tech program. Designed by our very own Fernando Serna, this program leverages the cutting-edge Azure Open AI Chat GPT 4.5 to create personalized career roadmaps for any role in the tech industry.
The Genesis of an Idea 💡
The inception of Starting in Tech stemmed from a simple question: how can we guide and inspire individuals looking to carve a path in the world of technology? The answer might be in leveraging the power of AI - more specifically, Azure OpenAI's GPT 4 - to dissect and understand the myriad of roles, skills, and pathways within the tech ecosystem.
How It Works: A Glimpse Under the Hood
The backbone of Starting in Tech is a sophisticated front-end interface built on the Wix platform, intricately designed to interact with our users. Here's a simplified overview of the magic that happens behind the scenes:
The AI-Driven Engine
At the heart of our program lies a series of backend functions that communicate with Azure OpenAI to fetch tailored information. These functions, including `processRoadmapSubmission` and `processJobSearchSubmission`, perform crucial tasks, such as:
Validation and Authorization: Using `wixCaptcha`, we ensure each submission is genuine, maintaining the integrity of our interactions.
AI Communication: Functions like `getChatCompletionStream` and `openAIJobSearch` are pivotal. They send user's data to Azure OpenAI, which then returns a wealth of information, from career advice to specific job listings, all tailored to each user's inputs.
Real-Time Updates and Engagement
Through `wix-realtime-frontend`, users see real-time updates on their dashboards. Whether it's the latest job postings or progress on their customized roadmap, this system ensures users are constantly engaged and informed.
A Closer Look at the Code
The process of leveraging Azure Open AI within the Starting in Tech program is a marvel of modern software engineering, encapsulating both the complexity of AI interaction and the simplicity of user interface design.
Let's delve deeper into the intricacies of how Azure Open AI instances are called and managed within this ambitious project. The focus here will be on highlighting the code snippets that directly interact with Azure Open AI, showcasing the seamless integration of AI-powered insights into the user's journey.
Azure OpenAI Call for Career Roadmap Generation
The creation of personalized career roadmaps involves querying Azure OpenAI with specific user inputs, such as their desired tech role and experience level. The getChatCompletionStream function orchestrates this interaction:
export async function getChatCompletionStream(channel, resourceId, user) {
// Setup for calling Azure OpenAI, including API endpoint and headers
const aoaiEndpoint = await getSecret('AZURE_OPENAI_ENDPOINT');
const secretKey = await getSecret('AZURE_API_KEY');
const deploymentId = "gpt-4-1106-preview"; // Example deployment ID
const endpoint = `${aoaiEndpoint}/openai/deployments/${deploymentId}/chat/completions?api-version=2023-12-01-preview`;
// Prepare the system message and user message for the API call
const systemMessage = systemInstructions[channel]; // Predefined instructions per channel
const userMessage = getUserMessage(channel, user); // Custom message based on user input
// The body of the POST request to Azure OpenAI
const body = {
messages: [
{ "role": "system", "content": systemMessage },
{ "role": "user", "content": userMessage }
],
stream: true
};
try {
// Axios client setup with OpenAI API key
const client = axios.create({
headers: {
'Content-Type': 'application/json',
'api-key': secretKey
}
});
// Sending the request and handling the stream response
const response = await client.post(endpoint, body, { responseType: 'stream' });
response.data.on('data', (chunk) => {
// Process each chunk of the response here...
});
} catch (error) {
console.error(`Error sending request to Azure OpenAI: ${error}`);
}
}
This function demonstrates the process of crafting a request that aligns with Azure OpenAI's expectations, including system messages that guide the AI on how to interpret the user's query and generate meaningful responses. It showcases the integration of dynamic content based on user input and the handling of streaming responses to provide real-time updates to users.
Beyond Code: The Impact
Starting in Tech isn't just about sophisticated code or AI algorithms; it's about the lives we aim to change. From the Site Reliability Engineer looking to deepen their expertise to the aspiring Software Developer making their first foray into coding, our program is here to light the way.
Special Thanks to The Wix Wiz: Eitan
As we bring this exploration of the Starting in Tech program to a conclusion, we must shine a spotlight on a pivotal figure who made much of this possible: Eitan, affectionately known as The Wix Wiz. Eitan's comprehensive guidance on integrating API calls with Wix has been nothing short of transformative. His tutorial, which you can explore here, offers not just a step-by-step guide but a deep dive into the possibilities that open up when Wix meets Azure Open AI. Thank you, Eitan, for being a beacon of knowledge and innovation in the tech community. Your contributions are truly making a difference.
This is just the beginning. With every roadmap generated and every career ignited, we move a step closer to our vision of a tech industry that is diverse, inclusive, and accessible to all.
Join us at SOMOS.tech and let's embark on this journey together! Your tech career starts now. https://somos.tech/start