Brad Ochocki

I'm a Data Scientist specializing in building end-to-end Machine Learning pipelines from scratch.

I write production-ready code that is readable, reliable, and scalable.


Experience

Data Scientist

Croptix | State College, PA

Croptix has developed a proprietary optical method that can look inside leaves to identify indicators of crop disease. This technology is field deployable, allowing growers to rapidly and economically screen large areas on a recurring basis, and manage diseases as early as possible.

I am Croptix's first Data Scientist, and have built our Python-based Analysis and Reporting codebase from the ground up.

My responsibilities include developing ML classification models for spectral data, building internal image processing and analysis tools, and constructing automated reporting pipelines.

I also manage the Data Team at Croptix.

NOV 2018 - Present

Data Scientist

MealPal | New York City, NY

MealPal is a subscription service that provides members with hundreds of lunch choices near where they live or work for a flat, monthly fee. Users can discover new restaurants and save money, while restaurants can increase revenue during their busiest hours of the day.

I was the first Data Scientist at MealPal, and built MealPal's first production ML model. The model made daily predictions about user actions across more than a dozen MealPal markets across the world, including North America, Europe, Australia, and Asia. The model immediately provided significant improvements in operating efficiency.

JUN 2018 - NOV 2018

Fellow

Insight Data Science | Seattle, WA

Insight is an intensive seven week post-doctoral training fellowship bridging the gap between academia and data science.

I created Skua, a Chrome extension that automatically filters out abusive Twitter content in real time.

JAN 2018 - MAY 2018

Data Engineer

Apex Systems | Houston, TX

Apex Systems specializes in providing IT professionals for contract, contract-to-hire and direct placements.

I built custom machine-learning software to integrate with client's in-house toolkit. The client's preferred development language was C#, so I could not rely on a robust, existing ML libraries and built most of the ML software from scratch.

SEP 2017 - DEC 2017

Graduate Researcher

In the U.S. alone, invasive species cost more than $100 billion annually in economic and environmental losses. These losses affect agriculture, forestry, fisheries, and transportation infrastructure.

In the U.S., invasive species cost more than $100 billion annually in economic losses, and affect agriculture, forestry, fisheries, and transportation infrastructure.

I studied how evolution can make invasive species spread faster over time, we can be improve predictions of invasive species spread.

My research generated high-profile publications, and was featured in Nature Research Highlights and the BBC World Service.

AUG 2012 - AUG 2017

R Programming Consultant

Simplify Healthcare provides niche software solutions and services to Health plans and third party administrator (TPAs), so they can navigate the rapidly changing market landscape.

As a short-term consultant with Simplify Healthcare, I re-built the functionality of legacy software for a client that provides market research and analytics to Fortune 500 companies.

OCT 2015 - NOV 2015

Energy Modeling Engineer

Vidaris | Norwalk, CT

Vidaris is a provider of niche consulting services within the architecture, engineering and construction industries focusing on high-performance buildings and specialty structures. They provide solutions for building envelope, energy efficiency, sustainability, code advisory and construction advisory projects.

I estimated energy cost savings by numerically simulating annual energy expenditure for multiple HVAC configurations.

OCT 2008 - JUL 2009

Education

Rice University

PhD, Ecology and Evolutionary Biology
2017

University of Nottingham

MRes, Evolutionary Biology
2011

University of Rochester

BS, Biomedical Engineering; Minor in Mechanical Engineering
2008

Skills

Programming Languages & Tools
  • Python (numpy, pandas, scikit-learn, Keras, package development)
  • R (tidyverse, RMarkdown, RStan, rjags)
  • Julia
  • SQL
  • MATLAB
  • C#
  • HTML
  • CSS
  • Git
  • Jupyter
  • Airflow
  • Flask
  • Periscope
  • LaTeX
  • QGIS
Techniques
  • Machine Learning
  • Building Analytics Pipelines
  • Timeseries Forecasting
  • Hierarchical Statistical Models (Bayesian and Frequentist)
  • Stochastic Simulation Models
  • Mathematical Models
  • Model Selection
  • Experimental Design
  • Image Processing