This project aims to develop and evaluate an AI-powered web application that provides tailored resume feedback for undergraduate students seeking internships and jobs. The goal is to enhance opportunities through scanning, keyword matching, and UI suggestions to help give resumes a clean, effective look. The web app enables students to upload or build a resume directly on the site, instantly receiving feedback from the AI powered assistant. Automatic formatting will help visualize the resume in real time, so it can be built in minutes. Resumes can then be instantly analyzed as they are being built, helping students quickly find positive patterns and enhance their resumes to the fullest extent.
Undergrad students often struggle finding jobs, especially in growing markets, such as computer science. Resume feedback often comes from recruiters who tend to rip apart a resume when they see any sign of inconsistencies with the standard. Resumes are online first now, and because of this, a digital analyzer is ideal. A web based analyzer could be used in any situation, even if it is right before an interview, to help fill in the gaps. Anyone who wishes to create a resume can do so without pulling up any software. Built on modern web technologies, such as React and Node.js, the tool can scan resumes for industry keywords known only to hiring managers and the assistant. The project implements a powerful OpenAI powered assistant that can help autocomplete skills or analyze a resume it is given. With these tools in one place users can decide to create a resume with AI or on their own and then analyze it, all while staying on the same website. A framework built around React will be used to develop the frontend of the website, handling all of the UI, and the Node.js framework will be used to implement backend processes. Together these two will communicate to send all relevant data to the frontend. A simple MongoDB database will be used to store resumes and user data. The backend will then communicate with the database as well as the OpenAI model that is being used for analysis. The same model will be used for analysis and building resumes as they require similar strengths, simplifying the OpenAI implementation. There will be a focus on UI to ensure all users quickly pick up on how to utilize the till naturally, reducing time wasted on learning a new system. An emphasis will be placed on speed and getting users results. After honing in prompts for each topic, the model will be fairly reliable and no further advancements will be required unless requested by the user base. Overall, there will be a focus on efficient, effective solutions to optimizing resumes for the workforce, prioritizing the everyday undergrad.
There will be completed in 6 stages: 1. Set up the basic frontend and backend boilerplate... Set up the basic frontend and backend boilerplate code and ensure they communicate with each other and the database. This requires each one to know the secret keys necessary to connect with their relative connections. The tools and frameworks used may change here, as they are updated regularly and may not align with previous, outdated frameworks they used to work perfectly with. This step is where all of these issues come up and get resolved. 2. After determining which tools fit best, the basic plans for each part are completed. First, a structure for the database is set up and refined so everything necessary can be stored. Once that is completed, backend function prototypes are defined, ensuring a user can do everything they would need to simply by using defined functions. Once this is complete a structure for pages can be created for frontend. With a focus on specific routing pages, methods, and data structures, frontend code can be fully planned. The next step here is to design the style for the frontend so styling can be done. Once this design is complete this step is finished. 3. Focus on the front end and setting up a basic, effective UI, where users can clearly navigate. This should mirror what was designed in the last step. This does not have any data loaded in and is specifically not a working website whatsoever. Instead it ensures the user gets a positive experience from the beginning. Every feature added goes through this so the user experience can be maintained. Frontend is completely tested to ensure each view accurately displays at all sizes of browser. 4. Backend is now connected to the database, ensuring the tables generated match what backend has described for them to ensure no bugs arise down the development road. 5. Frontend and backend are connected, ensuring they maintain their types and methods are references accurately. Each method is written and fully debugged. Frontend and backend should be in full communication. This completes development of the application. 6. Finally, tests are done using unit testing, focusing on edge cases and the general user to ensure every user hits no roadblocks.
The main deliverable is a functional, polished web-based resume analyzer and builder, deployed online. The product will be showcased through the URL and a presentation showcasing each feature and the development cycle. Each aspect of the site will be demonstrated with logs of unit tests. Backend code will be presented through PostMan or SwaggerHub. The main outcome, however, will be the site as a whole. This may also be presented alongside the SEO score to showcase the effectiveness, accessibility, and efficiency of the site. To ensure everyone has access to the site, these will all be perfect. After the stages of development, every bug should be removed and cleaned out, and features added will be forced through the same cycle to maintain reliability. Students will be able to use the website right away, potentially signing up with school emails to ensure the AI capabilities are not abused. Any undergrad student will have the opportunity to utilize the AI analyzer and builder, helping them through the hurdles of effective resume building. Keyword scanners will help students hone their resumes in and target specific fields or internships. If someone is in the process of applying, the analyzer can be used to help them determine their weak points and lean into their strengths when talking to recruiters. As with any modern website, one outcome is the potential for growth. Features specifically targeting technical interviews can be added to allow students to practice what they feel they are struggling with. Overall, the product would help many undergraduate students find internships and careers and feel more confident in their field. Students can have at least the weight of worrying about their resume lifted, as the focused, strategic, and effective AI assistant takes it from them. In one site, many students will be able to focus their time and effort on their more important skills required for executing well in the internship the product helps them acquire.
I have completed multiple web development projects including a portfolio website, live multiplayer games using WebSockets, and a mobile app using Flutter. I also developed software similar to the resume analyzer featured in this proposal during a software engineering course, building both the frontend and backend components. I’ve worked with modern frameworks like React, Next.js, and Node.js and have deployed multiple projects on Vercel and Render. These experiences have equipped me with the skills to produce this project effectively and successfully. I have worked with many different frameworks and technologies and feel confident that I could tackle this project and complete it within a semester. I have also worked with OpenAI APIs separately from these frameworks and could definitely combine my two areas of knowledge. While the project is substantial for a single developer, it is doable and completely in reach of anyone who has developed web applications.
This project does not require IRB or IACUC approval, as no human or animal subjects are involved.
This project requires no funding beyond standard personal development tools. All services used (Vercel, GitHub, OpenAI credits) are free or under free-tier usage.