Josephine Ndumu Email & Phone Number

DV Consult, Inc, MD: 2019 - PresentCloud Solutions Architect | Machine Learning | DevOps Engineer•Design CI/CD Pipelines for continuous integration, delivery and deployment using CodeCommit, CodeBuild, TestQA for manual approval within/ between stages, CodeBuild and CodeDeploy. Implemented CodePipeline to deploy applications on EC2 instances, lambda and ECS which led to faster delivery of services and retain customer retention by 20%. New features being released at periodic intervals, quick bug fixes, spontaneous response to feedback attracted users to products that led to improved loyalty and higher level of user satisfaction by 30%. CI/CD encourages transparency and accountability for problem amongst the team members; faster delivery to meet business goals.• Developed, managed, and maintained automated tools to simplify, standardize, and automate common repetitive administration, support, and scaling tasks using system manager Automation run book (run EC2 instances, stop EC2 instances, Terminate EC2 instances, Update CloudFormation Stacks (perform disruptive tasks in bulk) and Update AMI. EventBridge Events acting as a trigger for automation to update System Manager (target). •Applied custom, AWS KMS encryption keys to encrypt data in parameter store, EBS volumes on EC2 instances to ensure data encrypted at rest. TLS/ SSL certificates to encrypt data in transit, Applied sg according to best practices to allow customers around the globe to securely be connected to back end applications located in private subnets.

• Trained and deployed different machine learning classifiers such as Logistic Regression, SVM, Decision Tress, Random Forest, and XGBoost , Time series on various datasets• Utilized word embedding and TF-IDF to capture semantic and syntactic similarity of medical text data analysis• Applied several statistical techniques, such as feature selection, dimension reduction, and shrinkage for anomaly detection, model building, and hypothesis testing.• Conducted exploratory data analysis, mining, data visualization, data Preprocessing on clinical data to determine novel biomarkers for the disease attributes through statistical and similarity metrics with visualization and correlation measures for cluster generation and comparison Malignant vs Benign tumors, Conducted and analyzed performance of various neural net architectures such as DNN, CNN, VGG on clinical image data to identify the cognitive impairment in elderly patients