Alex Geiger
Looking to expand your staff? Then you're in the right place.
Determined, hard working with proven experience, and a passion for Data Science. Searching for a company looking to do big things with long term job stability. I would like to thank you sincerely for taking the time to visit my website. I very much look forward to an opportunity to speak with you further regarding future employment opportunity.
Phone: 585-626-2965
Email: ajg1444@rit.edu
About Me
Introduction
I would like to introduce myself as a candidate for the Data Science position within your company. I have a strong background with proven experience in natural language processing, sentiment analysis, Numerical methods, applied statistics and machine learning.
On a personal level, I possess a dynamic work ethic, along with an ability to think outside the box and pick up new skills quickly. I think commercially always, embrace positive change, am transparent and open with others.
I see this job opportunity to demonstrate and further perfect my craft as a Data Scientists and bring value to your company. In closing, I would like to thank you sincerely for taking the time to visit my website. I very much look forward to an opportunity to speak with you further regarding future employment opportunity.
Skills
Software Engineering
Python Java, Matlab, SQL, JavaScript R C++
Hadoop, Spark, Amazon EC2, Cassandra
Neural Networks, Bayesian networks
Data Structures, MapReduce, Pandas
Feature Engineering
Exploratory Data Analysis
Factor Analysis, Principal Component Analysis
Factor Analysis, Principal Component Analysis
Clustering, Multi-Class, Auto-encoders
Big Projects
Financial Market Modeling Through Deep Learning Neural Networks & High-Dimensional Data Embedding
Multi-agent, Deep learning neural network, and object-oriented program. The program optimizes portfolio return by analyzing data and modeling financial markets. Core concepts include MapReduce, Distributed Systems Exploratory Factor Analysis, & Principal Component Analysis. The aggregate data used in the project was collected via the Bloomberg terminal API.
Automated Inventory & Accounting Management Cloud-Based Solution
Provides a cloud-based inventory and account management solution. The program is in accordance with GAAP inventory management policies including LIFO and FIFO. Off-site data warehousing is compatible with both SQL and AWS data management systems and services.
Digit Recognition: Multiclass
Implemented linear classifiers to recognize handwritten digits from images. The dataset utilized the MNIST dataset. The multi-class solution involved using a deep convolution network which worked by applying each new image all pairwise classifiers and performing a majority vote to decide the output digit. k-fold Cross-validation was used to evaluate model performance. The calculated performance metrics of the different folds were combined and aggregated to a data structure.
The Maze: Heuristic
Single agent, state space, iterative A-star admissible heuristic search, object-oriented program. The program included six classes, Environment, Agent, A-star, Greedy, Depth, Breadth. A Euclidean distance state space was generated from the maze such that every element had a 1x1 unit size. The class A-star evaluation function used a heuristic to avoid expanding paths unnecessary paths.
Fraud Detection: Sentiment Analysis
Multi-agent, object-oriented program which Identified fraud in companies. Features were derived from sentiment analysis of 10-K filings MD&A sections and the quantitative analysis of cash flow, unexplained audit fees.
Courses that Changed my Life
Artificial Intelligence Explorations: 647
A graduate course which covered, features, agents, search, heuristics, logic, Bayesian networks, machine learning, reinforcement learning, deep learning and more.
In the course, I laid the groundwork for a deep learning network which predicts how to optimally pre-processed raw data then derive features and build a deep learning neural network tailored to how the raw data was processed.
Numerical Methods of Finance II: 736
The course provides students with a comprehensive understanding of mathematical and statistical models used in finance and their assumptions and limitations. Topics include delta hedging, Ito calculus, interest rate models and Monte Carlo simulations and more. A strong background such as linear algebra probability theory, numerical methods, and stochastics.
Data Science Projects
Learn About Our API
empowering research at low price
Access our data through a REST API powered by amazon gateway. Our prices are affordable the monthly payment is 1$ per month + amazon fees.
Why so cheap? The website is used to apply to jobs. To subscribe to the API email me for more information.
Notable Projects
Notable Projects
A different kind of resume
Come check out interesting projects I have worked on and the data behind them. These projects include forecasting:
Business Development
Company Deleveraging's
ACS Census Data
Company Performance
The data used in the projects was collected from a variety of sources including the American Community Survey Census, Bloomberg terminal API, Education Department Google Places API, and more.
Modern UI Toolbox
Create a dashboard
Bring your data to life with an interactive data visualization experience. Drag n' drop to plot geographic and time series data. Create modern tables and controls providing your users an enjoyable experience.
The toolbox provides an elegant solution to many of the problems with Matlab's current GUI and APP interface components.
