Team video
video here : https://www.youtube.com/watch?v=kWJJ5IK_7_M
Inspiration
We are a team of foreigners leaving in Finland. Having a family about there a need to transfer money regularly abroad. Why not provide them with the best exchange rates, meaning executing payments in the most favourable time. Analysing financial markets is a time-consuming and tedious task, so a personal bot could help with that.
What it does
The bot uses financial technical analysis like SMA (simple moving average), and (RSI) reference strength index to alert the user on a good time to execute a payment based on the frequency set during the configuration. For example, if the user wants to do the transfer once a week, a more reactive version of SMA is used, instead of a more stable SMA.
How we built it
We have tried out TransferWise APIs, connected Messenger APIs to the dotnetcore, added MongoDB data storage for caching user preferences, created a background worker for the running scheduled requests.
Challenges we ran into
TransferWise APIs are made with a test environment, so not all requests were working as expected.
Accomplishments that we're proud of
We got the bot and simple architecture working in such a short time.
What we learned
It was probably the best hackathon we participated in. We learnt a lot about financial markets, bots architecture, AI.
What's next for TransferBuddy
We believe that human language processing and bots are the future. The next for TransferBuddy is to execute commands as a real personal assistant. If we had more time, we would add more user flows, and conversation data, to make bot behave like a human.
Log in or sign up for Devpost to join the conversation.