Timeseries - Predictive maintainance system

Bangalore, India, 2017

  • Implemented an unsupervised machine learning algorithm for time series data on Spark, HDFS, and MongoDB
  • Engineered a real-time prediction model on Spark Streaming and Kafka, that scaled the performance by 100X
  • Collaborated with a team of 3, developed a web interface to monitor and train the system using Python, AngularJS2