Data Engineering,Analytics Engineering,Data Platform Engineering, and Machine Learning Engineering at scale using cloud

* Building data pipelines at scale using AWS EMR, Glue, EC2 and Snowflake data warehouse.* Application orchestration using AWS Step Function using compute via Lambda and Glue ETL.* Server less data pipeline using AWS services - Glue & Lambda and Snowflake tasks, pipes & streams.* Setup CI/CD pipelines and automate unit-testing in data application for reliable delivery and releases.* IAM roles and policy creation using CloudFormation via integration with JIRA, BitBucket and AWS CodeDeploy.* Gave FOUR tech talks about - git, Jenkins, Apache HUDI and AWS Stepfunctions.

Deep Learning Application to Understand Human Physiology. Worked under (Prof) Dr Richard Gault.

1. Bearing anomaly detection for trains in time-series data from mechatronic sensors using unsupervised machine learning.2. In-depth research with advanced signal processing methods - Fourier transform, wavelet decomposition, Discrete Wavelet Transform(DWT) and Continuous Wavelet Transform (CWT), spectrogram analysis, spline interpolation.3. Design and implement a pre-processing stage to enable data with heavy noise and distortion to give useful information regarding the faults in the train.4. Computer programming done using Python, C++, MATLAB. Query optimization to ingest data from InfluxDB in batches.

1. Developed an ASP.NET web application for real-time emotion detection and engagement tracker to generate feedback for students and course facilitators in the online learning community.2. The backend developed using Python and Convolutional Neural Network (CNN) implemented from scratch using TensorFlow and OpenCV.3. The front end of the web app communicates with the back-end via REST API.4. Training and testing data generated from scratch.5. High accuracy and performance of the model very efficient. Data augmentation used to train model using less data.