A web application that automatically finds easy-to-understand patterns, anomalies, and visualizations from csv datasets.

In this URL [login to view URL], there is a shitty prototype of the system that is incomplete but gives you an idea of what we want to achieve. I have the algorithms for mining the patterns, finding anomalíes, and creating the visualizations. Hence, I'm looking for a software developer that implements a SaaS application where users upload their data and the system finds and shows the patterns, anomalies, and data visualizations. The system will have a heavy workload in the backend because finding the patterns, anomalies and data visualizations is time-consuming. Hence, it must include a scheduler for maintaining a large number of jobs to execute. The system must tolerate the connection of many users simultaneously who sends their dataset; so the system must process all these datasets concurrently. The system includes many validations of the CSV file and the users get feedback when any error appears in those CSV files. The programmer must use TDD and must show us all the automatic tests properly working in every aspect of the system. The system includes paid plans for different kinds of users. Each plan includes some features such as, the number of data processing that the user can do in a certain amount of time, the length of the dataset, the cost for additional rows and columns, among many others. The system allows the user to keep a certain number of datasets processed according to the paid plan; and the user may delete some of the processed datasets. Also, the system allows the user to processed the same dataset many times but changing some parameters in the system. The system computes the patterns and anomalies using a proprietary console application created in C#; while the visualizations are generated with python code. Hence, the backend must execute these two things in order to process the datasets provided by the users. Some times the system finds tons of visualizations from a dataset, the system must load in the front end, only when the user requires some (hence avoiding loading too many things in the front end). The backend runs on a Windows OS.