[Music] inspiration behind the project i was inspired to create this mini project by the many global challenges the world is currently facing economic crisis wars climate change and more all of these issues either partially or entirely affect employment and career related matters my personal journey i myself have recently started exploring new opportunities and realized how complex and important this process can be that's why the idea of developing an application that helps people find the best opportunities for themselves match the right job to the right candidate seamlessly switch careers using their existing experience get guidance on choosing their future profession seemed incredibly relevant and exciting to me project mission i want to make this process faster simpler more accessible for anyone looking for employment or career growth in this constantly changing world one installing packages to get started make sure you have all the necessary packages installed run the following command in your project directory two downloading the API key to access external services and enable smart features in the application you'll need to obtain an API key set up your API key to run the following cell your API key must be stored it in a Kaggle secret named Google API key if you don't already have an API key you can grab one from AI Studio you can find detailed instructions in the docs to make the key available through Kaggle Secrets choose secrets from the add-ons menu and follow the instructions to add your key or enable it for this notebook if you received an error response along the lines of no user secrets exist for kernel ID then you need to add your API key via add-ons secrets and enable it three importing Google generative AI to use the power of Google's Generative AI in your app you need to install the appropriate SDK and import it into your code install the SDK if you haven't already install the Google Generative AI SDK four download résumés PDF the application supports downloading one or multiple user résumés in PDF format this feature is useful for recruiters career adviserss or the users themselves fifth generation of the embedding function the embedding function is responsible for transforming raw input data eg résumés job descriptions into vector representations allowing for better matching similarity search and recommendations how it works input data eg text or structured data is passed to the embedding function the function transforms the input into a vector representation the generated vector is stored in a database or used for comparison this material was mastered as part of the day two embeddings and similarity scores six working with Chromob chromb is a powerful tool for managing and querying vector data it provides efficient storage and retrieval of embeddings making it ideal for use cases involving machine learning similarity searches and other AIdriven applications in this section we will explore how to integrate and work with ChromB to optimize data handling and improve query performance in various projects this material was mastered as part of the day 2 document Q and A with R A7 matching candidates with the request the purpose of this step is to match candidates with the relevant requests eg job openings project requirements based on their skills experience and qualifications this feature will be particularly valuable for HR professionals as it allows them to quickly find the right candidate for a specific request by automating the matching process HR teams can save time and ensure that the best candidates are selected for each job or project reducing manual effort and speeding up the hiring or assignment process eight preparing a request for career analysis with Google Gen AI in this step we will prepare a request for career analysis using Google Generative AI this process involves gathering essential information about a candidate structuring it into a request and leveraging AI to provide insights on career development skill gaps job opportunities and career path suggestions this material was mastered as part of the day one prompting nine skills description and configuration for the request in this step we will define and configure the skills necessary for the career analysis request these skills will be used by Google generative AI to provide accurate career advice identify skill gaps and match candidates with relevant job opportunities this material was mastered as part of the day one evaluation and structured output 10 uploading old vacancies to create categories for finding related professions this step involves uploading old vacancies records to your system to create categories based on job titles these categories will help you to match the most relevant job titles with your CV to understand how to build a new prompt to further search for the most relevant jobs based on your CV 11th visualization categories in this step we will visualize job categories based on old or fake job records these categories will be created based on job titles helping you to better understand how to match job titles with your CV the visualization of these categories will allow you to generate more accurate prompts for future job searches ensuring the results align with your qualifications and career goals 12th generation of the embedding function in this step we will generate an embedding function that converts job related data such as job titles descriptions and required skills into vector representations these embeddings will help your system understand the semantic meaning of different job records which is crucial for matching jobs with the most relevant candidates what is an embedding an embedding is a mathematical representation of data where similar items are represented by similar vectors in a multi-dimensional space in the case of job titles and descriptions an embedding function will convert the text into numerical vectors making it easier for the system to process and compare different job titles based on their relevance to each other 13 displaying candidates skills and match job categories in this step we will display the candidates skills alongside the job categories that match the skills based on a specific request this will help visualize how the skills align with available job opportunities code explanation import JSON the JSON library is used to parse the response text containing the candidates skills extract skills we extract the list of skills from the JSON data and store them in a variable called skills extract job categories the job categories from a data frame of vacancies are stored in the labels variable display candidates skills the list of skills is printed in a well organized format showing each skill in a structured way display match job categories the job categories are displayed below the skills to demonstrate the possible job opportunities that match the candidates's abilities 14 analysis of similarities between skills and categories in this step we analyze the similarities between the candidates skills and available job categories by calculating the similarity between skills and categories we can determine which categories most closely match the skills a candidate possesses what is similarity analysis similarity analysis involves comparing two sets of data in this case candidate skills and job categories to determine how alike they are this process typically uses mathematical techniques such as cosine similarity to measure the closeness of two vectors representing the data in the context of job matching the goal is to find the job categories that are most relevant to a candidate based on their skills 15 search for jobs by selected categories this script allows you to search for job openings in Ukraine based on selected categories the job listings are retrieved using a model that searches for realistic job opportunities here's how the process works steps define job listing structure we define the structure of each job listing using pedantic models set up job search configuration a search configuration is created to request job listings in the desired categories search jobs by categories using the provided categories a request is sent to the AI model to search for job openings in Ukraine output results the job listings are then processed and printed this material was mastered as part of the day four Google search grounding 16 search for the most relevant jobs this task allows you to search for the most relevant job openings based on selected categories and skills the search is powered by AI which analyzes the similarity between skills and job categories to suggest the most fitting job opportunities steps define skill and category embeddings skill and category embeddings are generated to represent the features of both skills and job categories calculate cosine similarity the cosine similarity between skills and categories is calculated to determine which categories are most relevant for each skill search for jobs a search is performed for jobs based on the most relevant categories identified from the cosine similarity calculation display results the most relevant job openings are retrieved and displayed this material was mastered as part of the day four Google search grounding 17 preparing for an interview personalized advice this guide will help you prepare for an interview by providing personalized advice based on your skills experience and the job position you're applying for by analyzing the alignment between your skills and the job requirements you'll receive tailored recommendations to help you effectively prepare for the interview steps skill assessment start by evaluating your skills and experience this will highlight your strengths and areas for improvement job analysis review the job requirements and compare them to your skills to generate specific recommendations personalized advice based on the analysis you'll receive personalized advice including key interview questions strategies to present your skills and areas to focus on for improvement this material was mastered as part of the day one prompting 18 job application email generator this function generates a personalized job application email using the Gemini 2.0 flash model based on the candidates's profile in the job description purpose automatically creates a professional tailored cover letter that aligns with the job requirements and showcases the candidates strengths returns a string containing the generated job application email ready to be sent conclusion: The system for searching and analyzing career opportunities is a powerful tool that significantly simplifies the job search process it not only analyzes the candidates skills but also helps optimize the interview preparation process saving time and resources for both candidates and employers of course these are just the minimal capabilities utilized with the help of Google Gen AI which can be applied to similar problems this system can be continuously improved and expanded as the possibilities are limitless while time is limited which is why I've decided to go ahead and start looking for a job [Music]