Parinith Rajkumar Email & Phone Number

Texas, United States

Dallas-Fort Worth Metroplex

San Francisco Bay Area

• Maintained a security-oriented common internal Java library for use by developers in AppDynamics that interfaces with AWS KMS, AWS Secrets Manager, and Hashicorp Vault.
• Consolidated a featureflag management Java library that implements LaunchDarkly SDKs improving usability and developer onboarding experience.
• Pioneered and deployed Spotify Backstage to expand developer experience through a unified dashboard connecting Jira, Confluence, Tea mCity, and custom products to a single point of contact.
• Operated multiple Gradle plugins to enable rapid integration with CI/CD, publishing, and deployment tools
• Monitored projects and plugins utilizing Splunk and created additional alerts to improve monitoring accuracy.

West Lafayette, Indiana, United States

• Collaborated with members from Caterpillar Inc to coordinate and deploy a predictive modeling pipeline that streamlined petabytes of data to predict the probability of breakage of parts within vehicles.
• Charted and executed a product development plan employing Agile methodologies.
• Supervised a team of students by delegating duties efficiently and mentored them in cloud technologies such as AWS

Reston, Virginia, United States

• Visualized and implemented a high-volume analytic pipeline initiative that ingests semi-structured text datasets and applies unsupervised hierarchical clustering to extract necessary data.
• Performed entity extraction, natural language processing, supervised and unsupervised topic analysis to identify and visualize driving features and textual themes over time in text datasets with 300K records.
• Partnered and optimized a new visualization method for datasets with more than 50k nodes that reduced the runtime by 10% implementing statistical analysis and machine learning.
• Quantified metrics based on points of interest and created a model with appropriate thresholds.

West Lafayette, Indiana, United States

• Automated manual preprocessing methods utilizing python decrease time consumption for data cleaning by a factor of 50% and decrease repetitiveness.
• Envisioned and refined a pipeline by applying the ggplot2 library to provide relational metrics on datasets with 500k+ records decreasing runtime by 15% and reducing CPU utilization.
• Incorporated additional quantitative and qualitative metrics and reasonable thresholds, improving the accuracy of the predictive analysis model by 10% in research attempting to predict animal food habits in a region.

Bachelor Of Science -Computer Science, Data Science

Bachelor Of Science - Bs, Applied Statistics