Data Scientist by day, Software Engineer by night. An open-source enthusiast, who enjoys designing, implementing and deploying machine learning models in production systems. I love working in multidisciplinary teams and can contribute equally to a research or industrial environment.
GPU accerelated Convolutional Neural Network library in ROOT, the data processing framework developed and used at CERN.
Designed and implemented a fully functional Convolutional Neural Network library in CUDA, which is currently integrated in ROOT, the data processing framework written by and used at CERN. The library allows the timely processing of particle physics experimental data. My implementation consistently outperforms the previous CPU version by a factor of 3, as its performance and memory footprint is comparable to those of Keras.
Design and implementation of Data Science solutions for the Dutch industry.
During my second year I have been working individually for the Dutch Public Prosecution Office (OM) in a long term project geared towards early detection of money laundering activities. The project's stakeholders also include the FIU, as well as a large financial institution.
During the first year I led and participated in multidisciplinary teams working on short term projects for our industrial partners.
I designed and implemented a monitoring and intervention capable system intended to facilitate remote control of the companies servers. This allowed developers to safely launch and kill test VMs without having to go through the system's administrator.
June 2016 - September 2016Engineer's Diploma (BSc and MSc combined studies) in Electrical and Computer Engineering, with a specialization in software engineering and machine learning (then called pattern recognition). Graduated among the top 10% of graduates. My thesis on accerelating genomics workflows using distributed computing environments, was graded with 10/10 and published at the 12th International Conference on Artificial Intelligence Applications and Innovations (AIAI 2016).
September 2010 - July 2016
I initiated and contributed to a team of colleagues participating in multiple machine learning competitions. I mentored team members in modern software engineering practices including object oriented programming and version control, by delivering on-site workshops, writing comprehensive guides and conducting code reviews.
Succesfully completed a series of four courses, as well as a capstone project in order to learn Scala and Spark, as well as the Functional Programming Paradigm. My capstone project was an application able to process and visualize large-scale temperature datasets. The source code is not published in accordance to coursera's honor code.
Integrated a new GUI into a robotics application written in C++. I also implemented several design and performance optimizations, one of which prevented a potential memory leak. The project is currently distributed as an official ROS package.
Data Scientist | Software Engineer
© 2018 steremma
