I built data-investors for my own personal investment research and selections, but have begun opening the site to the public due to frequent requests.The core site’s backend is built using Python (Django), gunicorn, nginx and AWS RDS, and is deployed via Docker Compose through an automated CI/CD pipeline (buildkite) to ec2. The docker images are stored in AWS ECR, while the static files, media files, and data files are all stored in s3.The site uses web scraping tasks (scheduled via Celery) to pull data at various intervals from across the web. I built the execution code of the tasks as a separate microservice deployed via Docker as an AWS Lambda Function, which allows me to scale up concurrency and compute power for large task loads while only paying for the compute time used.After the data is scraped, additional tasks extract and compress the relevant information. Multiple data pipelines (built using DBT) then act on a Redshift Cluster to clean and organize the data into various datasets for training and evaluating predictive models. The pipelines run on a regular cadence and output datasets to s3 (compressed) for further use, allowing me to turn my Redshift cluster off while not in use ($$$!).From a modeling perspective, I have built a series of Django apps to train, evaluate, compare, and productionize models based on each model’s target. This consists of selecting an algorithm, which features to use, which subset of data to use, how much historical data to include, model hyperparameters, and feature preprocessing methods. A series of tasks run on the newly-created models to backtest each over a period of at least 2 years. Finally, a ‘Model Builder’ app analyzes all models trained to date and, based on historical improvements, creates new model configurations for evaluation, repeating the cycle.

Led experimentation and analytics around social interactions and experiences, including designing and implementing A/B tests to measure the incremental lift of new product features and analyzing experiment results to recommended ship/no-ship decisions.

As the instructor for Rutgers University's Online Data Science Bootcamp, I lead a group of 30+ students over 24 weeks to dive deep into the core skills required of data professionals today, including Python Programming, Databases and ETL Processes, Front-End Web Visualization, Data Visualization in Tableau, Fundamental Statistics, and Machine Learning (Supervised, Unsupervised, Deep).

Utilized artificial intelligence to build the company’s internal retention model including research, development, planning, implementation, and productionization of the model – Model utilized in evaluation of all strategic business decisions including: market launch decisions, pricing decisions, capital raise decisions, hiring decisions, underwriting decisions, and capital structuring decisions.Leveraged custom built Retention Model to run conterfactual forecast scenarios, including forecasting retention and revenues under different pricing plan changes, different rate-capping strategies, and different product-mix triaging strategies. Utilized the price change scenario output to create a simplified spreadsheet-based elasticity and revenue forecasting tool for use by state management in evaluating potential rate change strategies and their expected impacts.Supervised and mentored an Associated Data Scientist and a Data Scientist I on work related to customer churn modeling, model feature analyses, and model monitoring and improvements. Onboarded and trained new hires on utilizing AWS Infrastructure.

As only the third person to join Root's lean finance team, I worked with the Chief Financial Officer in managing all aspects of financial reporting, budgeting, forecasting, FP&A and regulatory updates. Built company's internal financial model with GAAP and Statutory financial statements, hundreds of adjustable inputs for scenario analyses, visual exhibits to quickly demonstrate the impact of changes on key KPIs, and reinsurance forecasting.

Cofounded company, created pitch book, built team, and secured funding. Raised over $4m in financing to establish and grow the company from accredited investors across the US. Completed application for an Ohio-Based medical marijuana company to process and sell cannabis-infused cookies and other medicinal products.

Managed all back office activities and employees, including operations, inventory management, finance, budgeting, forecasting, financial planning and analysis, human resources and compliance.

- Conduct proprietary and assigned research on equity and fixed income investments for retail investors using SEC filings, Morningstar Reports and Orion- Summarize daily research, economic reports, and mutual fund conference calls into quick hit summaries, utilized by broker during client interactions- Compile quarterly reports for clients with broker to update and advise on investments

- Aid in the processing and payment of financial obligations of The Ohio State University in a timely and accurate manner- Work alongside other employees to maintain file organization and inter-departmental communication in relation to financial obligations

- Researched Natural Killer Cell development in Acute Myeloid Leukemia - Trained in advanced biomedical research techniques, including hemocytometry, cell culture, real time quantitative polymerase chain reaction, flow cytometry, RNA isolation, cDNA production, data analysis and western blotting.

Created, updated and maintained a file management system for the law firm.

Researched alcoholism in a wistar rat model. After exposing rats to alcohol for a 10 week period and verifying dependence, treated rats with Ceftriaxone to activate Glutamate Transporter-1. Doing so increased GLT-1 Production, thus increasing the glutamate reuptake and attenuating alcohol cravings.