MUTO - (ZQV1, NAAR)

Empowering Pathways: Jobs, Mentors, and Resources for a Fresh Start.

Comment 1

Inspiration

We were inspired by Future of Education, as well as the drive that some of these vulnerable individuals had. We've all been on transit, and 2 of our members including me ended up having a good conversation and realized that all they need sometimes is guidance and acceptance.

What it does

We use a voice analysis and database in order to connect vulnerable populations to mentors, to get hands on learning into trades and other services, instead of just odd and labor jobs, to fully integrate them back into society and providing more workers. Our website is mainly the service that connects it all together, being a centralized platform for both life resources and career resources. Using ML we were able to find the most common words and do emotional analysis as well to check compatability

How we built it

We originally started brainstorming and designing on figma, and we spent a particularly long time on the pipeline, since we wanted this to be as welcoming and easy to use since this was more those inexperienced with tech, or at least not perfectly versed. We then built out the website using Node.js frameworks, and lots and lots of JSON files to start with, until we got a database up and running which allows us to give personalized recommendations based on an individuals stories

Challenges we ran into

The biggest challenge was getting our machine learning model to work properly, especially since the h5 from python is not compatible with javascript, the JSON conversion is something we did externally and then used mainly for testing and finding common words that we could use for anger, sadness, and other sentiment analyses.

Accomplishments that we're proud of

We are proud of the fact that in the 24 hours we actually got market validation and people from the real world to give advice to our idea. In addition to this we were able to implement majority of the features for the most part

What we learned

We learned that tackling too many features at once is very stressful, even with a large team, being able to implement one or 2 good features at a time brings ease of mind more so that half-finishing 5 good features. We also realized that while front end solutions do work, a good backend just makes the entire process of front end smoother, almost like instant vs delayed gratification

What's next for MUTO

While we do have a machine learning model, we believe that a web scraper would also be very useful at finding resources, so it's not limited to manual input. This would also involve automation in other areas such as using maybe indeed or service API's for trades, and partnering with institutions to offer online course content, even though it may be difficult for vulnerable individuals to access

Built With

Share this project:

Updates