Data Engineer

Michael Devin Floyd’s Master Resume

Email: *******@*****.*** LinkedIn: michaelfloydms

Phone: 713-***-**** GitHub: mdf3039

Current Position(s):

• President of Transfyr

• Mentor/Reviewer for Udacity’s Data Scientist, Machine Learning Engineering, and Self- Driving Car Engineering Nanodegrees

Skills

Top Three Skills: Problem Solving(Advanced), Critical Thinking(Advanced), Coding(Advanced) Programming Languages: Python, SAS, R, C++, Java, SQL, IronPython, JavaScript, XML, WinBUGS, C#, XAML, CUDA, Swift

Applicable Software: Alteryx, Tableau, Spotfire, BayesiaLab, PLINK, KING Big Data Tools & Amazon Web Services: Spark, Hive, Hadoop, EC2 Instances, Lambda, Cloudwatch, EMR, S3, RDS, DynamoDB, VPC, API Gateway, SES ML Algorithms & DL Platforms: Decision Trees, SVMs, NNs, CNNs, Recursive Feature Elimination, Gaussian Mixture Models, AdaBoost, PCA, Tensorflow, Keras, Theano Experience

Mentor at Udacity (March 2018 – current)

• Mentoring students in cohorts for the Machine Learning Engineering, Data Scientist, and Self-Driving Car Nanodegree programs.

President of Transfyr (January 2018 – current)

• Front-End Developer, Back-End Developer, and Designer of the mobile app Transfyr Graduate Research Assistant at Clemson University (August 2018 – December 2018)

• Postulating and researching ideas and algorithms tailored to advance the field of Biomedical Data Science and Informatics (BDSI)

• Senator Ambassador for BDSI program at Clemson University Data Scientist at DeepIQ http://deepiq.co/index.html# (October 2016 – January 2018)

• Wrote code that processes raw data from numerous sensors on dozens of oil rigs around the world. Used change point analysis to obtain state of each rig per second.

• Classified rig states into sections that produce key performance indicators that highlight quality of rig operations. The final output will be broadcasted and used by rig operators and managers to discern the efficiency of operations.

• Combined Machine Learning and Deep Learning algorithms on 2008 PHM data to predict the remaining useful life of turbine engines. Used an ensemble to separate the predictable features and predict the failure of the engines based on statistics from those features. The results showed the model had incredible precision and accuracy predicting the remaining useful life when the engine failed within 20 cycles.

• Coded/Used a Multi-Scale Convolutional Neural Network with an additional branch for autoregressive tendencies. Amazon Web Services (AWS) instances were used to train the network. The goal was to observe if Convolutional Neural Networks, which are usually associated with picture processing, can be used to find/observe patterns in multi- dimensional time series data. This did not prove to generate better results than the machine learning approach.

• Using AWS, read, formatted, and preprocessed LAS formatted files into EMR stored in Hadoop's distributed file system, and used for future Hive querying. This allowed for processing and understanding the well logging files.

• Created various interactive visualization demos with Spotfire using processed and simulated data to visualize the performance of rig operations and predict performance of future rigs' completion time, bit failures, and locations. Created data functions and used scripting for hard-coded changes. The purpose of each demo was to demonstrate to clients our capabilities to deliver a great platform and highlight analytical skills. Tutor at Wyzant (December 2013 – October 2016)

• Tutored students in St. Louis, MO; Lafayette, LA; and Baton Rouge, LA using wyzant.com, helping in disciplines such as Mathematics, Physics, and Chemistry. Teacher at Jefferson Parish Public School System (January 2016 – May 2016)

• Taught Mathematics at Grace King High School in Metairie, LA. Research Assistant at Washington University Division of Biostatistics, (September 2014 – December 2015)

• Researched and studied genetic predisposition in lineage studies to perform data analysis and quality control for a renal study in Salt Lake City, UT

• Analyzed genotype, phenotype, family, and pedigree data

• Filtered exome chip data and produced final exome dataset for the statisticians across the country to evaluate and analyze.

Biostatistician Intern at Washington University, Institute of Public Health, (May 2015 – August 2015)

• Implemented a parametric survival curve to medical expenditure data

• Created new method to bootstrap large-scale complex data through researching and reading methods about bootstrapping and processing large-scale complex data efficiently. This allowed the combination of years of complex data taken by the MEPS and confidence intervals obtained from each parameter in the survival curve. Student Worker at LSU Math Department and LSU Recruiting Services (June 2007 – May 2011)

• Coordinated campus tours for incoming and prospective students. Assisted freshmen students scheduling classes. Administered make-up tests Education

Degrees:

• MS in Biostatistics, Washington University in St. Louis, GPA: 3.65

• BS in Mathematics, University of Louisiana at Lafayette Nanodegrees:

• Predictive Analytics for Business Nanodegree Program, Udacity

• Self-Driving Car Engineer Nanodegree Program, Udacity

• Machine Learning Engineer Nanodegree Program, Udacity Presentations

Presentations: Tutorial on ‘Introduction to Spatial Analysis through Statistics’ given at the Conference on Statistical Practice (2017); Poster Presentation of the ‘Complex Sampled Bag of Little Bootstraps’ at the Conference on Statistical Practice (2016)

Contact this candidate