Frontier is a no-code AI-driven analytics platform developed by Fluid AI to create an end-to-end solution for analytics and AI-based problems.
I collaborated with my colleague Mihir Deo during my tenure as a Data Scientist at Fluid AI on the development of the product. Some of the key features developed are
Data Optimization
- This feature generates a validation report using the current data, including the results of multiple statistical tests, best practices, and recommendations.
- All the report recommendations can be implemented in the data using a single button from the GUI.
Model Optimization
- Multiple levels of model optimization were integrated into the product.
- A GUI to modify the default hyperparameters of the model was implemented.
- An advanced feature was implemented to find the best hyperparameters for a given data along with the configuration to define the best hyperparameters.
- An automated model feature was implemented to find the best model for a given data along with the configuration to define the best model.
Model Monitoring
- Alerting features were implemented to notify when model maintenance or retraining is required based on statistical analysis.
- Dashboards are built that show the intrinsic details of drift in data.