Rajagopal Subramanian

Title: Rajagopal Subramanian: Machine Learning and Data Processing Innovations

Introduction

Rajagopal Subramanian is an inventor based in Plano, Texas, whose patent work focuses on machine-learning systems and efficient data processing. His listed inventions address ways to standardize model development, organize reusable data features, and streamline the training and deployment of machine-learning models. He is associated with Humana Inc. and has 2 patents.

Latest Patents

Subramanian’s latest patents include “Machine learning platform and feature generation for efficient data processing” and “Machine learning platform and pipeline for efficient data processing.” Both describe systems designed to support agile model development and reduce redundant work by coordinating and standardizing model training and deployment.

The inventions describe obtaining data, preparing it for feature generation, and reformatting and de-sensitizing it for storage. Features are kept in locations accessible to models and training modules, allowing them to be reused without repeatedly adjusting data for each new model. To develop a model, the system establishes an evaluation cohort and identifies a model template and relevant features. These are used for experimentation and evaluation, after which training artifacts, such as model weights, are recorded in a model store. Model scripts and settings can then be registered in a centralized database for execution.

Career Highlights

Subramanian’s patent portfolio reflects an emphasis on infrastructure for machine learning rather than a single model application. The described systems bring data preparation, feature reuse, experimentation, model training, and deployment into a coordinated workflow. This approach is intended to help data scientists develop and operationalize models more efficiently.

Collaborations

Subramanian is associated with Humana Inc. His listed coworkers include Keegan Nesbitt and David Christopher Mack.

Conclusion

Rajagopal Subramanian’s patents describe reusable data features and standardized machine-learning workflows that aim to make model development and deployment more efficient.