Project Info
Inspiration
It’s becoming increasingly common for new college graduates to be seeking their first job. Even if they have been building up their experience, applying for their first job is a very daunting task. It can be mentally draining and discouraging. Post-grad life? Radio silence. You're expected to navigate job applications, networking, and career pivots, all while everyone around you seems to have it figured out.
What it does
GradPath helps to fix this problem. It helps to create a community with like-minded individuals navigating the post-grad scene and combines this community bonding with AI-tailored recommendations powered by ToolHouse AI, which helps to give people real, actionable steps based on relevant information, such as their major or goals. We also have utilized vector-based peer matching powered by ChromaDB in order to find these like-minded individuals who have the same goals, same challenges, and same journey. Not random networking but precision matching. Users who match with others will have the option to either join a community chat or a 1-on-1 direct chat with one of their best matches.
How we built it
GradPath is a full-stack powered platform in the minds of current college/newly graduated students. Users are suggested tasks via a query to the ToolHouse API, which has our custom-designed agent tasked to develop bigger goals into smaller subtasks, organizing them by priority. Having smaller subgoals will make bigger tasks much more doable. Our peer-to-peer matching system is powered by an LLM retrieval pipeline that extracts peers with the most similar background based on cosine similarity representations of their goals and interests. Our front-end is powered by React, while our backend is powered by Python while FastAPI allows for effective communication in between.
Challenges we ran into
Our biggest challenge was debugging the frontend architecture as we were faced with tedious errors on our ambitiously built platform. We also had issues with the website displaying information from the backend properly.
Accomplishments we're proud of
We implemented an LLM-powered strong user-matching framework. We are also proud of being able to build an interactive UI that includes many small animation effects.
What we learned
We also adapted to big challenges that came along and to sudden changes outside our control.
What's next
Utilize different APIs to address scalability as the number of users grows. In addition to our matching pipeline, we want to give the peers the ability to direct message each other.
This project did not link a GitHub repository.
Analysis
No indexed repository for this project, so there are no commit stats to show.
Technology
- CSSUnchecked
- FastAPIUnchecked
- JavaScriptUnchecked
- PythonUnchecked
- ReactUnchecked
- Tailwind CSSUnchecked
No repository was indexed for this project, so these Devpost claims have not been checked against code.
AI coding agents
No repository was indexed, so agent usage could not be checked.
Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.
Codebase size
No repository was indexed, so there is no codebase to measure.
This project did not link a GitHub repository, so there is nothing to diagram.
This project did not link a GitHub repository, so its feature claims have not been checked against code.
Export this project's context (description, README, evidence, key source files) to chat with an AI agent elsewhere.