Project Info

LifePuzzle

This project did not submit a demo video on Devpost.

Inspiration

As students still figuring out how to plan our lives, it can be hard to break down our goals into something manageable. We wanted to create a program for people like us who need a little extra help working towards our goals, and on a daily basis.

What it does

LifePuzzle will take a goal given by a user, and break it into various tasks. LifePuzzle will give you one task a day, and as you complete your tasks, it will give you puzzle pieces until you have an entire puzzle to represent your progress.

How we built it

We utilized Together AI’s API to generate tasks for the user each day based on the goal they input. We also made it a bot for Discord, because it is already populated with a very similar user base, marking it easy for us to get users, and making it very convenient for them to use our bot.

Challenges we ran into

It was very hard for us to figure out what medium we wanted to present our program in. At first we considered a website or make it its own standalone chatbot, but we felt that those options did not encourage easy and convenient use, nor were they easy to develop.

Accomplishments we're proud of

We are really proud of the overall presentation! The bot is able to provide meaningful and beneficial tasks that align with the user's goals, and the UI is also very easy for users to get the hang of.

What we learned

We learned how to use Generative AI in our projects in order to implement functionality that would've been hard/impossible to implement otherwise. Together AI's API was extremely easy to use for this project, and we hope to utilize it more even after the hackathon.

What's next

We would love to work on it more to give it more features and possibly make it a Duolingo-like mobile app. We would also like to create a better reward system for it to give users even more incentives to complete their tasks.

calhacks-transfer-app

App to help community college students figure out their course plan, regardless of where they want to go.

Utilizing:

MindsDB for data storage & preliminary data clean up with AI
together.ai for AI data processing on the user's end
??? for frontend

Idea

Background: John is a CS student at De Anza College (a community college), who wishes to transfer to a university in the future. he's a very ambitious student, and wishes to go to either a UC, an Ivy League college, or MIT.

Since John is studying at a California Community College, figuring out his plan for the UCs is easy! He just has to go on assist.org, enter his college and select one of the UCs, enter his major, download the articulation agreement, copy down the classes that the articulation agreement tells him to take, and repeat this whole process all over again for each UC campus. How easy!

For Ivy League and MIT, it's even easier! He just has to dig through countless Google searches to eventually find a page from a specific college, denoting what they're generally looking for in transfer applicants, then compare their criteria to the different classes his college offers, then repeat again for the next college.

There is an easier way to do this. Enter [PLACEHOLDER NAME].

With [PLACEHOLDER NAME], it can:

automatically analyze the assist.org articulation agreement for multiple colleges
provide students with an easy way to figure out the transfer requirements/criteria of out-of-state/private colleges
merge all of the information it has into one simple and easy to read page, with Generative AI being able to further clean up the information to make it more human-digestable.

Data Handling for Individual Universities

DB Structure:

classes_requirements (each row is a class requirement)
uni_name - University Name - For convenience's sake
uni_id - University ID - We either generate this ourselves or we borrow from assist.org whenever possible
data_type - Source of the Data - e.g. "california_assist", "private_scraped", "public_scraped"
major - Major this data point is for - Can be specific (e.g. "Computer Science") or general if it applies to all majors (GE reqs)
class_code - A specific class needed by the school - Needs to be the specific course code required by the school
class_name - Name of the class

[TODO] Add a row for AI to process

california_assist (each row is a class equivalency between two schools)
sending_name - Sender College Name
sending_id - Sender College ID
receive_name - Receiving College Name
receive_id - Receiving College ID
sending_course_code - Sender College Class Code
sending_course_name - Sender College Class Name
receiving_course_code - Receiving College Class Code
receiving_course_name - Receiving College Class Name

Analysis

Compare with all teams

View

Metric

RyanC
15

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only

PythonIn code

1 of 1 appear in the indexed code.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

24 KB

Source files

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.