SyllAib.us

Inspiration

Every year, nearly 65% of students graduate from college feeling completely blindsided by the job market and unprepared for their professional lives. Not because they lacked ability or effort, but because they lacked direction.

“Why am I doing this?” and “What am I doing wrong?” aren’t just expressions of frustration, but rather indications of a larger issue: we’ve built a system where education and career preparation are treated as two separate paths. A system that fails students from all over the nation.

What it does

We didn’t create a resume builder. This isn't another LinkedIn. This is the first time a student can look at their course load and see their career reflected back at them– personalized, specific, and actionable from day one.

Through SyllAib.us, we aimed to capture the dynamics and constantly shifting priorities of students’ learning journeys, and how every step in it, down to the courses chosen to be taken, affects the opportunities students qualify for and are presented with in the future. It turns a static syllabus into an interactive, data-driven career roadmap that guides students to be more intentional in their current decisions to reskill and learn, based on their current skills and career goals.

The experience starts with the user simply uploading a PDF of the syllabus to the system. SyllAib.us then parses important logistical information–professor’s office hours, grading breakdowns, the # of quizzes and exams to be expected within a semester–and displays it in a clean, easy-to-manage dashboard. Multiple PDFs can be uploaded, with all of ur courses being presented and available to you at your convenience.

But SyllAib.us isn’t just a syllabus parser. By entering your major and career goals on your dashboard, the system will then act as a mentor to develop the most optimized roadmap to get to where you want to be, from where you are now, beautifully visualized in the form of a knowledge map.

SyllAib.us constantly updates your strengths and areas of concern, the more it finds out about you. By utilizing APIs, the dashboard also displays targeted internships and occupations that users can pursue, no matter where they are in their learning journey.

How we built it

We were intentional in building our pipeline for the agent to aggregate data to the user, specific to their goals. We were careful in balancing the incorporation of Large Language Models that generated text with API calls that pulled factual information, to make sure the data pipeline presented the user with relevant information.

We built the app using the NextJS framework to create a full-stack application using Node as backend and React as frontend, and we connected MongoDB to the application. We were careful to make sure the neatness and minimality of the UI reflected the recluse the app intended to be, where the user can calm and collect themselves.

Challenges we ran into

Everything.

Kidding! But there definitely were a ton of roadblocks along the way. Some challenges include:

Database set up:

Identifying the data that had to be stored in the database, and the schema that allowed for the best data traversal and transformation
Setting up a MongoDB database storing user information, connecting it with the frontend

APIs:

Figuring out how to use new APIs
Bypassing the rate limits posed by the different APIs that we had integrated

Accomplishments that we're proud of

Made an experience personalized to every user, and carefully chose features that were central to the app’s function, connecting the user’s academic journey to their foundational career goals.
Created a full-stack application with full routing capabilities
Designed a beautiful ui for users, with individual sessions for each user
Visualized a knowledge map that users can interact with to see potential curriculum maps, generated based on your goals
Established a MongoDB and a fully working relational database with no experience
Used OAuth successfully to log in with GitHub and Google
Created a full registration and login page using MongoDB and REST API

What we learned

Deepti: How to properly utilize multiple APIs to enhance the functionality and efficiency of our product.
Sania: Balancing data flow pipelines that balance concrete user data to minimize LLM hallucination and provide useful and relevant information to the user
Nik: Learned login sessions, authentication, relational data modeling, and mapping data to UI
Dash: Learned how to implement MongoDB and connect it to a NextJS project

What's next for SyllabAi?

Scaling. We asked around--this is a product that a lot of the students we talked to seem to love and would use in their daily lives. To accommodate such traffic, we'd have to work to scale our product, something we'd love to talk ;)

A few more things we'd implement in the future: