ASPIRE

• 

  Poster

Inspiration

The inspiration behind Aspire arose from the real-world problem where students struggle to make informed decisions about their academic and career pathways due to a lack of personalized guidance. We recognized that students cannot systematically identify their academic strengths and weaknesses solely from historical performance data, and there is no data-driven mechanism to match these academic profiles with actual industry requirements. This results in inefficient academic planning and a misalignment between educational choices and professional aspirations, motivating us to build a solution to bridge this gap.

What it does

Aspire is an AI-based academic recommendation system comprising three integrated modules: automated academic history parsing, LLM-based career matching, and intelligent course recommendations. Practically, the system allows users to upload their transcripts (as PDFs or images), automatically extracts grade data and performance trends to analyze student strengths. Based on this profile analysis, Aspire provides comprehensive career recommendations complete with salary estimates, required skills, and growth prospects, while also suggesting relevant courses to close the student's skill gaps.

How we built it

We built Aspire using Next.js as the primary framework for full-stack development, supported by MongoDB for storing user data and transcript history. For document processing, we integrated Cloudinary for file management and utilized pdf2json along with OpenAI Vision to extract raw text from uploaded transcript files. The core intelligence of the system is powered by the OpenAI API, where we engineered complex prompts to analyze transcript text data and generate structured JSON outputs containing detailed and measurable career recommendations.

Challenges we ran into

One of the biggest challenges we faced was handling the variety of unstructured academic transcript formats, both in PDF and image forms. Ensuring accurate text extraction from these documents was crucial for valid AI analysis, requiring us to implement robust text cleaning logic to remove unnecessary characters and handle token limits. Additionally, crafting the right prompts so that the LLM consistently returned data in a valid JSON format without hallucinations required extensive iteration and testing.

Accomplishments that we're proud of

We are incredibly proud to have created an end-to-end system that transforms static academic documents into a dynamic and interactive career roadmap. The system's ability to not only read grade data but also provide context such as "Day in the Life," career ladders, and industry growth metrics for every recommendation is a significant achievement. Furthermore, the seamless integration between visual data extraction (OCR) and deep textual analysis proves that this technology can serve as a reliable digital academic counselor for students.

What we learned

Throughout this project, we learned a great deal about the importance of robust data handling when working with LLMs. We realized that the quality of AI output is heavily dependent on how clean and structured the input data is, driving us to deepen our data preprocessing techniques. We also learned how to balance giving specific instructions to the AI while allowing the model to use its "creativity" to provide relevant yet factually grounded advice in an academic context.

What's next for Aspire

The next step for Aspire is to expand the database and capabilities of the course recommendation system to be more precise in closing user skill gaps. We also plan to improve the accuracy of the parsing module to support more academic document formats from various educational institutions. The goal is to make Aspire a holistic platform that not only advises but also facilitates the student's learning journey from start to career readiness.

Built With

• cloudinary
• mongodb
• next.js
• openai