Legacy Academic & Career Advisory

Inspiration

We wanted to build a tool that was able to advise students on their future plans since it can be overwhelming to start thinking about.

What it does

This Flask application uses AI tools to analyze multiple data sets with information about alumni and current students. The application asks the student about their current academic journey and standing and helps to recommend courses and activities based on what other students in their field have done.

How we built it

For our framework, we used a Python Flask application for the backend and a combination of JavaScript, CSS, and HTML. To utilize DoIT's provided database, we used TigerData for our database. We used Gemini's Generative AI and Microsoft Copilot to build most of the project structure and imported the api into our project to create customized responses and comparisons. Furthermore, we used the ElevenLabs API to create text-to-speech on the fly given Gemini's personalized output.

Challenges we ran into

One of the biggest challenges we ran into was debugging, especially when using generative AI. We spent most of our time trying to understand what the code being generated was doing, and how to move forward to our goals.

Accomplishments that we're proud of

Natalie hand-drew each avatar for each section of the project; she made the complete design, and we believe this will set us apart from the competition

What we learned

For one, we learned Solana was used with crypto wallets. But beyond that, we learned about how the use of API's and combining them together can create a more powerful product. especially with the way we put Gemini and Eleven Labs together to create a more interactive and wholesome experience.

What's next for Legacy Academic & Career Advisory

Firstly, we need to reduce the amount of hardcoding that occurred when Gemini was writing code for the project so that the application can be dynamic and responsive if the database changes. Next, we need a truly accurate dataset. We believe that when we have an accurate dataset that the true audience can relate to, we will be able to evaluate our project better. It's not a perfect project, and it needs some tweaks. We would especially like to increase the use of the Gemini API, so the experience is more personal and feels unique to you. However, the limit of API requests makes this task difficult for now. Another thing we would want to do is add functionality for advisors and staff to use the product to better suit students' needs.

Built With

• copilot
• css
• digitalocean
• elevenlabs
• flask
• gemini
• godaddy
• html
• javascript
• python
• tigerdata