Daniel Evans Email & Phone Number

New York, Ny, Us

New York City, Ny, Us

New York, Ny, Us

- Trained, tested and deployed a survival model in Python (using scikit-learn and lifelines) that predicts average time to conversion for users exposed to display ads for a major insurance company by implementing the Cox proportional hazard algorithm.- Mitigated media teams’ reporting needs by creating automated dashboards in Looker, powered by data from Google Search Ads 360, Google Search Console, and Adobe Analytics.- Collaborated with data engineering team to ETL proprietary client sales data combined with Google Analytics user data into a central datalake for the purpose of making in-depth analyses that track the full consumer journey from display exposure to sale.

New York, Ny, Us

- Developed a KPI that measured the conversion rate of users exposed to display channels before submitting a lead by examining consumer journeys from CM360 data transfer logs on BigQuery.- Generalized the overlap analysis from pairwise comparisons to n-fold comparisons across multiple categories of interest.- Formulated a mixed media model using a random forest regressor with Shapley value interpretation to describe the contribution of each media channel to total lead generation and sales.

New York, New York, Us

- Ingested 20+ years of historical securities data to a SQL relational database for a major financial law firm- Automated ETL process for over 35k flat files with mapping routines developed with Python and regex- Saved thousands of potential man-hours in the discovery process for Section 10(b), 11, and 14(a) lawsuits - Spearheaded analytical efforts to improve employee performance for a major IT firm from its financial and operations data

San Francisco, Ca, Us

- Developed BikeFlow (cbalancer.site), which predicts bike activity to and from a given Citibike station to identify hotspots for rebalancing efforts using publicly available trip data- Deployed a random forest regressor with scikit-learn to predict bike traffic flow from various categorical and continuous input features- Munged and blended multiple sources of data to a backend PostgreSQL database using psycopg2 and pandas- Hosted a website on AWS developed in Flask with gunicorn/NGINX

New York, Ny, Us

- Contributed to the understanding of colloidal stability with direct force measurements of charged particles in suspension using optical traps and high speed imaging- Improved the precision of colloidal interaction measurements by a factor of 10–100 over atomic force microscopy by using optimal statistical methods and artifact-free particle tracking- Implemented supervised learning algorithms to evaluate candidate mathematical models for an underlying physical system- Developed custom software in IDL to detect image features, perform statistical analyses, and visualize findings from gigabytes of particle interaction videos

Chicago, Il, Us

- Investigated the liquid-like properties of granular materials by examining their behavior under the influence of electric fields- Employed feature detection and tracking algorithms in IDL to extract data from high-speed videos- Derived aggregate physical quantities from the analysis of particle motion

Doctor Of Philosophy (Ph.D.), Physics, Soft Condensed Matter Physics

Bachelor Of Arts (B.A.), Physics

Physics, Mathematics