CashGraphs is a fintech web app for running modern portfolio theory optimizations. Select a "flavor" to specify your investable asset universe, a risk metric, and any weight constraints, and then run the optimizer. The algorithm will pick out a full efficient frontier of optimized portfolios for all risk preferences.
Anyone who manages securities portfolios or is involved in the asset allocation process.
Frontend is Vue 2 + Vuetify. There is no styling beyond the Vuetify defaults because I know nothing about CSS. Charts are done with ChartJS and very little customization. The backend algorithm is an AWS Lambda function that runs a Torch optimizer to return the efficient frontier chart plot points and portfolio weights.
I'm a personal financial adviser, and I like getting into the weeds on asset allocation stuff. I also like learning coding stuff and thought that portfolio optimization was a cool use case for machine learning optimizers like TensorFlow and Torch (the first few versions of the backend used TensorFlow). I use this optimizer for my client portfolios, and I thought that larger-scale portfolio managers might be interested in it as well.