Professional Experience

Designed and developed a framework as a micro service to observe machine learning models in production and also analyze their performance using various statistical tests.
Devised the framework such that data scientists/engineers were able to observe the performance of the model in production environment using a single button click.
Contributed in building a microservice to re-train a deployed model on demand.

Collaborated with an agile team to launch an in-house anomaly detection system to assess end-to-end data and application reliability clusters hosted on AWS using Pyspark, Apache oozie and Cassandra.
Analyzed the collected time series data and real time insights were visualized using REST APIs and Grafana in the form of a dashboard which helped the QA team to ensure data accuracy and availability by increasing failure detection speed by 100% (4hrs+ to ~real time)

Built and maintained monitoring applications on Hadoop for data integrity checks using python which helped the QA team to monitor the production failures in real time using an alerting mechanism.
Orchestrated CI/CD activities to deploy on AWS using automated python scripts, git and deployment tools like Jenkins, Ansible. Developed pipeline was used by dev and QA team for automated rollout of more frequent releases with ease.

Created generic UI testing framework to create a test resource for a particular page, run selenium tests based on the configuration provided during the test resource creation.