Goldman's MarketMinds

• 

  Landing Page

• 

  Market Performance Tool

• 

  Coming Soon!

What Inspired us

We wanted a safe, simple, cheap way to predict a companies value for a consumer wishing to invest in a long-term portfolio option without having to do too much research. We can use the public information of senate and congress disclosure on stocks to evaluate whether a company's earnings will raise or lower. Senators and congress people have teams of advisors and usually invest at large scales so why not follow their lead.

How we built Goldman's MarketMinds

We built our project in Flask with python and next.js. Our AI models were built with TensorFlow in python and trained with carefully treated and massaged data.

Challenges we faced

Most of the projects had to do with artificial intelligence, machine learning and data science which we all have little experience with. We chose to not settle for a full-stack web development project as we could all do that. To learn we have to push ourselves.

Accomplishments that we're proud of

Most of all, we built and trained an AI/ML model that achieves our goal in estimating company earnings. Our team worked tirelessly to collect and massage the data to train the model and then use to get predictions.

What we learned

We learned how to properly clean and transform data to prepare it for data visualization and metric evaluation. This data was used to train our AI/ML model which was new to us.

What's next for Goldman's MarketMinds

We would want more metrics to have weight on the prediction for company earnings such as sentiment from social media and news articles, searches for the company and other similar metrics. Consumers will be able to weigh the metrics they perceive are the most important as they wish to achieve their goals.

Built With

• chart.js
• figma
• github
• java
• javascript
• next.js
• python
• react.js
• springboot
• tensorflow