Phone +1-424-***-**** firstname.lastname@example.org www.linkedin.com/in/george-terzakis
UCLA ANDERSON SCHOOL OF MANAGEMENT Los Angeles, CA Master of Financial Engineering December 2019
Focus: Data Science and Machine Learning, Statistical Modeling, Behavioral Finance, Financial Innovation UNIVERSITY OF PIRAEUS Athens, Greece
Bachelor of Engineering in Computer Science July 2018 Thesis for senior year project entitled as “Data Envelopment Analysis - Applications in Banking” Selected coursework:
UCLA: Machine Learning and statistical modeling, Econometrics, Time Series Analysis, Behavioral Finance Online courses: Python for Data Science and Machine Learning, Deep Learning - Artificial Neural Networks, SQL, Hadoop SKILLS
• Programming and Scripting: Python, R, SQL, Tableau, Hadoop
• Libraries: numpy, pandas, scikit-learn, tensorflow, keras, pytorch, pyspark, matplotlib, seaborn, plotly EXPERIENCE
LOS ANGELES CAPITAL Los Angeles, CA
Applied Finance Project (Master Thesis): “Text Based Signals using NLP” July 2019 - December 2019
• Researched an Alpha opportunity by identifying a novel source of text data
• Utilized Natural Language Processing to target and evaluate the customer’s sentiment towards products listed on Amazon for small-to-mid sized firms.
• Explored how different sentiment techniques (Bag-of-words, Polarity function) compare to each other in terms of effectiveness based on their accuracy
MOTOR OIL HELLAS Athens, Greece
Data Scientist Intern June 2019 - September 2019
• Built an ETL pipeline to store the data related to clients’ payments and defaults provided from the subsidiary companies (SSIS, SQL Server)
• Developed a machine learning model to assess client creditworthiness
• Translated the business problem into a methodological problem, in which we want to estimate the probability of default for each client. Trained a Logistic Regression using historical data of the prior clients along with macro- economic factors and performed cross-validation to assess the performance of the model and achieved 0.87 accuracy.
• Automated procedures using VBA and Excel which assisted in preparing the company’s quarterly report 10-Q BLOODE (Online blood donor registry, nonprofit Startup) Athens, Greece Founding Member April 2016 - May 2018
Event Coordination & Management
• Initiated corporate and government collaborations resulting in $30,000 sponsorships and donations
• Coordinated public and private events for groups of 50 to 1000 people to promote blood donor awareness
• Managed company growth of the company from 3 to 20 staff members; recruited, trained and supervised 15 volunteers
• Created an Artificial Neural Network to solve a Customer Churn problem
- Used a sample of 10,000 customers of a bank and their personal credit data to create a demographic centric model which predicts which of the customers are at highest risk of leaving the bank (0.8625 Accuracy of Model)
• Implemented a Recurrent Neural Network to predict Stock Prices
- Trained a stacked LSTM model on five years of Google’s stock price (2012-2016) and successfully predicted the upward and downward trends of the future stock price.
• Experimented with Boltzmann Machines and Autoencoders to create a Recommendation System
- Built two recommendation systems, one Boltzmann Machine, to predict whether a user will like or not a specific movie based on previous ratings on a number of movies and achieved an accuracy 0.75 and one Stacked Auto- Encoder which tries to predict the rating (1-5) of a movie by a user.