About Me
Hi, my name’s Harisyam and I’m a Data Scientist @ INNIO Jenbacher (previously GE Power). I mainly do forecasting with Deep Learning using LSTM’s, I also have experience in data engineering. I really love working in an interactive team.
I am always motivated to work in various areas of Data science field that will challenge me continuously and allow me to use my education, skills and past experiences in a way that is mutually beneficial to myself and my employer and allow for future growth and career advancement
I am most skilled in Python and very familiar with AWS Technologies
Check out my CV: here
Projects
- I started this project to learn the concepts of Data Warehousing using AWS Redshift to basically support the projects at my work
- I enjoyed the Capstone project from Udacity where all the concepts from ETL are thoroughly discussed and were implemented in this project using the Data Orchestration tool Apache Airflow
- I have also learnt data processing using PySpark Livy API in this project
- Developed a robust Machine Learning pipeline for dealing with noise and skewness in very large databases
- Implemented a custom Deep Neural Network model comprising of denoising autoencoders in PyTorch with python which can deal with the presence of annotation and feature noise together with skewness in very large data sets
- Developed a GUI for process automation of NVH-CAE testing for DAIMLER Trucks using python Qt framework
- Key Achievements: Various softwares are used by the DAIMLER Trucks Big Data Team for analyzing the root causes of truck damages when used in semi-autonomous mode
Experience
R&D Project: Development of a scalable analytics pipeline for classifying vehicles with risk of catalyst failure Customer: FORD Motors, US
Reinforcement Learning based Thermal Management for Cabin Heating Mode Selection Funded by: ECSEL FRACTAL, EU Commission
- Implementation of reinforcement learning based models aimed at improving energy efficiency and reduction of environmental pollutants by having effective heating mode selection for Cabin Thermal Management
INNIO Power(previously GE Jenbacher GmBH)
https://www.innio.com/enData Scientist
August 2018 - July 2021
Analyzing Terabyte scale data to deliver deeper insights and find anomalies across the entire fleet of 20K Gas Engines
- Developed Financial analytics and created Tableau Dashboards for analyzing various revenue streams of the company
- Created entitlement forecasts for predicting the customer demands of various engine spare parts for the forthcoming quarters
- Analyzed and investigated different Root Causes (RCA’s) for catastrophic Engine Failures using time-series anomaly detection methods f rom python’s sklearn library
- Developed a Full-Stack Deep Learning-based solution for maintaining the Emission sub-system in the engines (Preventive Analytics)
- Developed various forecasting algorithms for calculating Remaining Useful Life of engine components using Variational Autoencoders (VAE) and LSTM’s developed in Tensorflow using python
- Key Achievements: Standardized automatic billing process across the company for the services revenue stream which had generated stable revenue even during the COVID-19 crisis
Education
RWTH Aachen
M.Sc. Computational Engineering
Oct 2015 - Mar 2018
During my time at RWTH I had been learning the application of computer science techniques in the field of mechanical enginnering. I had spent most of time developing algorithms for the automotive lab and also teaching MATLAB for various students
BITS Pilani
M.Sc. Chemistry and B.E Mechanical Engineering
Aug 2010 - July 2015
A Little More About Me
Alongside my interests in Data Science my hobbies are:
- Bicycling
- Volley Ball
- Running
- Bouldering