Summary
Overview
Work History
Education
Skills
Timeline
Generic

Andrew Zirm

Reisterstown

Summary

A highly analytical and adaptable individual with excellent communication and teamwork skills. Eager to contribute to team success as a member of management. Positive attitude and committed to continuous learning and growth. Confident in ability to thrive in new environments with new challenges.

Overview

12
12
years of professional experience

Work History

Analytics Engineer / Senior Data Engineer II

Greenhouse Software
04.2021 - Current
  • Member of the AI Ethics Committee, contributing to our vision for responsible AI and help us navigate our legal obligations while maintaining our ability to innovate with new technology
  • We had been using a vendor, datacoral, for our ETL pipelines and data publication needs
  • That vendor was no longer suiting our needs and eventually went out of business
  • In 2021-22 I worked to make the datacoral-based pipelines more reliable and cost effective, e.g., leveraging the still-new Redshift Serverless to produce customer-facing and data-driven presentation decks in an on-demand and scalable manner
  • We have been running dbt on Snowflake for more than a year now
  • The Snowflake costs month-over-month have been steadily increasing
  • I led a project to implement a new architecture for our key Greenhouse Recruiting pipeline
  • This new method allows us to cut costs by limiting the amount of data needed to be processed when it runs
  • It also consolidates the logic for the pipeline in a single stored procedure which enables any optimizations we make there to apply to all the tables running in the pipeline
  • As part of the migration from Datacoral to dbt+Snowflake: I led the effort to replace the data publication capabilities using Hightouch and planning the effort to build a home-grown data publication solution
  • Our Business Intelligence Connector product (BIC), now runs on Snowflake
  • I gathered the requirements, architected a solution, and led the project to implement this key customer-facing pipeline on our new tool stack
  • Member of the Continuous Education Cabal, trying to make sure all Greenhouse employees continue to learn and grow

Senior Data Scientist II

Greenhouse Software
10.2019 - 04.2021
  • Resume parsing helps our recruiting platform be more friendly to incoming candidates, filing out forms based on the parsed data rather than having to fill it all in again
  • I helped during the vetting process prior to Greenhouse acquiring the company, CV and Me, which uses the latest LLMs to parse resumes into structured data and supports new features such as resume anonymization

Senior Data Scientist

Greenhouse Software
04.2019 - 10.2019

Data Science Mentor

Thinkful and SharpestMinds
09.2018 - 08.2022

Data Scientist III

Greenhouse Software
01.2018 - 04.2019
  • Greenhouse wanted to provide our users with insights into the state of their job pipelines
  • Using historical data pulled from our data warehouse, we built a predictive (Bayesian) model to calculate the next hire date for each open role on the platform
  • This is powering the 'GH Predicts' feature which launched to most of our customers in 2018

Data Scientist

Greenhouse Software
10.2014 - 01.2018
  • Greenhouse wanted more visibility into our data
  • I wrote custom ETLs (Python/Airflow and Clojure) to get data from various sources and populate a Postgres data warehouse
  • Then, using BI tools such as Periscope Data and Mode Analytics, I built dozens of dashboards including: Data from product usage, customer support tickets, net promoter score and customer metadata from Salesforce to build a churn and engagement model in Python/scikit-learn
  • Tracking recruiting KPIs for our customers
  • These metrics and charts are now a key part of our Customer Success Managers' quarterly business reviews

Data Science Fellow

Insight Data Science
08.2014 - 10.2014
  • Developed a web app to help expecting parents efficiently explore the space of possible baby names using the name meanings, history and usages (using Python)
  • Used Python+NLTK to create a term-document matrix
  • Each row is a name and columns are the top 5000 words and bi-grams from the name documents
  • Used Python+Flask and Twitter Bootstrap to build the front-end

Associate Professor

Dark Cosmology Centre
08.2012 - 08.2014
  • Developed a new course, Classic Papers in Astrophysics, for MSc and PhD students
  • Built statistical models and data pipelines for a large galaxy imaging survey
  • Advised four PhD students (who have all graduated successfully now)
  • Conducted research using the Hubble Space Telescope and Very Large Telescope

Data Science Mentor

Insight Data Science
10.2014 - 10.2016
  • Led SQL seminars for groups of ~30 Fellows
  • Interviewed Fellows on variety of topics including machine learning and data preprocessing
  • Led consulting engagements with two Fellows over two sessions of Insight

Education

Ph.D. - Astrophysics

Johns Hopkins University
Baltimore, MD

Skills

  • Python programming
  • Data pipeline design and management
  • SQL
  • BI tools
  • Statistical modeling and machine learning
  • Critical thinking
  • Effective communication
  • Continuous improvement

Timeline

Analytics Engineer / Senior Data Engineer II

Greenhouse Software
04.2021 - Current

Senior Data Scientist II

Greenhouse Software
10.2019 - 04.2021

Senior Data Scientist

Greenhouse Software
04.2019 - 10.2019

Data Science Mentor

Thinkful and SharpestMinds
09.2018 - 08.2022

Data Scientist III

Greenhouse Software
01.2018 - 04.2019

Data Scientist

Greenhouse Software
10.2014 - 01.2018

Data Science Mentor

Insight Data Science
10.2014 - 10.2016

Data Science Fellow

Insight Data Science
08.2014 - 10.2014

Associate Professor

Dark Cosmology Centre
08.2012 - 08.2014

Ph.D. - Astrophysics

Johns Hopkins University
Andrew Zirm