Visiting Physicist @ SLAC National Accelerator Laboratory
Bachelor of Science (BS) @
University of Wisconsin-Madison
I am a scientist by nature and passion. I enjoy learning about learning and my current position as a Data Scientist at Brighterion allows me to do just that. On one day I am learning how to utilize new and novel machine learning techniques and applying them towards detecting fraud in the payments industry and the next
I am a scientist by nature and passion. I enjoy learning about learning and my current position as a Data Scientist at Brighterion allows me to do just that. On one day I am learning how to utilize new and novel machine learning techniques and applying them towards detecting fraud in the payments industry and the next I am doing an exploratory data analysis concerning data in the financial and equities industry.
Currently I am interested in the data visualization of exploratory data analysis, i.e., how best to communicate the results of an analysis to to non-technical audiences. The work that I do with payments data is used to mitigate fraud and deals with banks bottom lines, consequently any type of results we produce need to be digestible by a business crowd in order to be considered. It's a big challenge but my experience presenting technical results to expert and non-expert audiences, gained during my PhD work, as well as my teaching experience, is a hugely valuable asset. There is a lot of great visualization software and I am excited to incorporate it for this necessity.
I also am interested in machine learning (ML) and its applications. At Brighterion we work with payment industry data where fraud is the number one concern. In order to prevent it we need to understand it before we can prevent it. This requires standard ML techniques such as data mining, Neural Networks and ensemble learning, but as fraud is constantly evolving there is a lot of room for innovation in this space. That could be in the form of new types of feature-engineering as well as both supervised and unsupervised learning.
Programming/Coding: Monte Carlo statistical models, data analysis. Proficient in C++, Python and SciPy, SQL and shell scripting.
Contact directly at andrew.ruland [at] gmail.com
Data Scientist @ In my role as a Data Scientist at Brighterion I am actively involved in developing machine learning (ML) and artificial intelligence (AI) solutions to fraud in the payments world. The models we create help some of the largest acquirers and banks in the world to detect and stop loss that occur in real-time. This involves using AI to learn about individual behavior and machine learning to detect deviations from that baseline. Additionally, fraud behavior is constantly evolving. Together, these types of behaviors require a supervised and unsupervised approach to help mitigate the effects of theft.
Some of my duties and responsibilities include:
- Creation of AI and ML solutions for fraud in payments industry. This includes data mining, MLP feed-forward neural networks and ensemble learning methods as well as Brighterion's in-house analytics suite, iPrevent.
- Research and Development exploring supervised (recurrent neural networks, auto-encoders) and unsupervised machine learning (density-based, k-means and nearest neighbor clustering) and feature-extraction strategies to improve existing fraud solutions as well as extending capabilities to other domains including financial market analysis.
- Lead data scientist on a project for potential clients demonstrating a business case for using Brighterion technology in a domain outside of e-commerce.
- Writing technical reports recommending strategies for data utilization in the payment industry to non-technical and business audiences.
The main tools I use to accomplish these wide range of tasks are SQL and bash scripting during the model building, and for the more exploratory analyses and research I have introduced python and the scipy stack into the company toolkit. The Brighterion predictive analytics suite, iPrevent, is also an invaluable tool when creating models. From 2014 to Present (1 year) San Francisco Bay AreaVisiting Physicist @ Measurement of partial branching fractions in rare B meson decays (BF ~ x10^-6) using maximum likelihood parameter estimation, machine learning with decision trees, Monte Carlo simulations. From September 2012 to July 2013 (11 months) Postdoctoral Research Associate @ Made precision measurements of rare B meson decays to probe physics beyond the standard model From June 2010 to August 2012 (2 years 3 months) Graduate Research Assistant @ Rare B decay searches From June 2005 to May 2010 (5 years) Austin, Texas AreaLaboratory Assistant @ Led discussion and experimental labs concerning kinetic and thermal physics to undergraduate pre-med major students. From August 2003 to December 2005 (2 years 5 months) Austin, Texas Area
Doctor of Philosophy (PhD), High Energy Physics @ The University of Texas at Austin From 2003 to 2010 Bachelor of Science (BS), Physics and Astronomy @ The University of Wisconsin at Madison From 1998 to 2003 Andrew Ruland is skilled in: Machine Learning, Data Analysis, C++, Programming, Python, R, Scientific Computing, Numerical Analysis, LaTeX, Matlab, Spectroscopy, Statistics, Shell Scripting, Linux, Unix
Looking for a different
Get an email address for anyone on LinkedIn with the ContactOut Chrome extension