Data scientist with a master’s in applied analytics from Columbia University. Military veteran with over six years of experience serving in leadership positions and directing highly trained teams of up to 30 personnel. Served in 12 countries on 4 continents. Currently a Consultant at Accenture Digital in the Applied Intelligence Group, Data Science Practice and Associate Professor at Columbia University.
- Python (ScikitLeran, Pandas, Numpy, TenserFlow, Keras,PySpark)
- AWS (S3, EMR) — Completed Solutions Architect Certification Course with A Cloud Guru
- TigerGraph Graph DB (GSQL)
- GSQL, SQL, Presto, Snowflake
- R (Rshiny, ggplot2, CARET, randomForrest, glmnet, ROCR, gbm, tree, dplyr, stringr, rjags, Rmarkdown)
- Familiar with Databricks(IDE Community Edition, MLFlow), Tableau, H20 Driverless AI
- MS Office, Project, Visio, Draw.io,JIRA
- MS – Columbia University, Applied Analytics Aug 12, 2018
- BA – University of Washington, Business Administration, magna cum laude
- Major in Finance and Minor in Economics Aug 18, 2017
Data Scientist – Xandr Feb 2020 – Present
- Identity (Tigergraph, Python)
- Generated queries to inspect KPI’s that will report on status of ID graph being generated with Tiger graph
- Created custom weighting algorithms for edges on community graphs
- Generated playbook template python notebooks to run EDA.
- Segmentation (Spark/Scala, Python)
- Developed age and gender inference model for devices based on mobile browsing behavioral data to help support identity resolution.
- Developed methodology and table structure for indexing algorithm to support faster segmentation on mobile browsing data.
- 1 patent submitted, 3 Patents pending submission as of 11/11/2020
Associate Professor – Columbia University Sep 2019 – Present
- Capstone (APAN 5900): Fall 2019, Summer 2020, Fall 2020
- Assisted lead professor in class management by issuing surveys, keeping attendance, facilitating discussions, and presenting course content.
- Evaluated student performance and guided solutions for delivery to client that met or exceeded requirements.
Consultant, Applied Intelligence, Data Science Practice – Accenture Nov 2018 – Feb 2020
- Audited models to support large scale expansions for international teams including model validation and recommendations for efficiency improvements.
- AT&T/Xandr: Identity Resolution and Management (Anaconda Enterprise, AWS,S3,EMR,Spark/Scala,RStudio,Atom)
- Built a logical framework to resolve identities across multiple data sources allowing for a more realistic implementation of their identity graph.
- Built an auto scaling framework to be able to expand audience segments for marketing campaigns using stacked ensemble techniques.
- Built an audience segment creation tool that would build organic segments based on behavioral data using KMeans.
- Created > 60 Models with Spark/Scala — RF, LR, Logistic Regression, SVM, Multilayer Perceptron — Used standalone models and stacking method. These models were used to predict various Demographic, Econographic and Psychographic traits to impute missing data upstream for identity resolution.
- Designed and created a KPI table that was used to build a dashboard for model management.
- Performed EDA on > 10 separate datasets ranging from 1-15GB in size.
- Generated POC for EDA tool to develop key client capabilities.
- HP Enterprise: EDA and Data Cleaning (SQL, SSMS, RStudio, and JupyterLab)(6-Weeks Full-time)
- Performed EDA and NLP pre-processing on millions of customer service requests to buildout a taxonomy of issues that would be used for future project solution design.
Columbia University – MS Applied Analytics Course Projects Aug 2017 – Aug 2018
Course Highlights: Machine Learning, Anomaly Detection, Bayesian Analysis, Research Design, and Data Visualization
- Utilizing NLP to analyze enforcement actions and identify risky financial institutions. (Accenture 150+ hours)
- Managed a team of 8. Analyzed bank behavior that could identify early warnings of regulatory violations.
- Collected over 2GB of both unstructured and structured data, performed exploratory data analysis (EDA).
- Using R to perform the analysis with a variety of ML libraries for regression, clustering, and classification.
- Built a shiny dashboard for final deliverable integrating data and results of from analysis.
- Created a model to predict expected returns of S&P500 stocks that included industry relationships. (40+ hours)
- Used 4 week returns for 10 years of 700 different stocks. Used R to clean and JAGS to run the model.
- The model used specific sampling techniques to identify the distributions of returns for each stock at each period. (MCMC method using a Bayesian Hierarchical Model)
- The model improved on the error of using a simple average of S&P500 returns, by 10% and the relationship was found to be significant.
- Developed an equity portfolio optimization framework that analyzed between 250-300 assets, rebalanced a portfolio in less than 3 minutes, improved rebalancing costs with 6 additional constraints to the traditional Markowitz problem. (150+ hours)
- Combined over 660MB of data including returns, industry ID, and covariance matrices.
- Created an algorithm to take the given benchmark data, create basic Markowitz model and integrate the constraints directly. Built unique functions to clean the data, format it into the requirements for our own unique optimization model. Used R exclusively and completed the project in 10 weeks. One of 28 teams who submitted a solution out of more than 160 participating teams.
Security Officer, Edmonds Community College (Full-Time) Dec 2014 – Aug 2017
- Maintain security and safety of the college campus with over 10,000 students and nearly 3,000 employees while providing services to staff and students to facilitate daily operations.
- Filed over 30 incident reports ranging from public disturbance to sexual assault and coordinated with local law enforcement for most reports.
Sergeant, United States Marine Corps (Full-Time) Jun 2007 – Apr 2014
- Lead a team of four to review operations efficiency in developing a new command and control system.
- Stress tested a new system with live training scenarios and thorough after actions analysis on requirements.
- Evaluated internal processes of the new system to determine potential efficiency improvements on current standards.
- Evaluated and advised over 50 subordinates on job performance and conduct.
- Held regular training classes to teach on various topics from military equipment and operations to testing on qualifications relating to job performance and advancement.
- Trained over 100 and certified over 30 Marines in the Marine Corps Martial Arts Program
- Mentored over 20 Marines on job performance and personal achievement.
- Created visual displays of operations for the Commanding Officer for an Air Wing of over 50,000 personnel.
- Created digital overlays of maps, symbols, and information that was updated in real-time.
- Created handmade overlays on printed maps to facilitate high mobility missions and support overall progress of the battlefield.
- Responsible for all financial operations and planning for events ranging from 50 to 430 personnel.
- Managed 3 accounts through regular auditing and transactions.
- Planned and executed monthly happy hour functions for embassy personnel and related partners.
- Planned the Annual Marine Corp Ball, a formal black-tie event, for 430 people including high ranking military and foreign diplomats.
- Stocked, and ran the Marine Store which started with 3 items and grew into 15 unique souvenirs.
- Managed and grew the Marine welfare fund from about $2,000 to over $15,000 to be spent on the future Marines.
- Developed and executed security plans for Ambassadors, Chiefs of Mission and Regional Security Officers.
- Conducted 7 drills per month of various security response scenarios.
- Responded to 10 actual security threats and performed after actions reports which lead to updating security plans.
- Regularly reviewed and updated various security response plans to incorporate threat trends and capabilities due to changing resources.
- Data Scientist, Xandr Feb 2020 – Present
- Associate Professor , Columbia Universit Sep 2019 – Present
- Consultant, Accenture Digital Nov 2018 – Feb 2020
- Campus Security Guard, Edmonds Community College Dec 2014 – Aug 2017
- Sergeant, United States Marine Corps Jun 2007 – Apr 2014
Certifications / Awards
- Top 3 in Kaggle competition, NFL 1st and Future Analytics (2020), Prize Award
- Data Scientist with R, DataCamp (2018)
- First place in NYC Open Data L-Train Innovation Challenge (2018) at Grand Central Tech
- Leadership Snohomish County Young Professionals Class (2016, Washington State)
- Navy Achievement Medal: Awarded for peak performance and proficiency (2014, Venezuela)