Summary
Overview
Work History
Education
Skills
Software
Certification
Honors
Publications
Timeline
BusinessDevelopmentManager
Duane M. Lee

Duane M. Lee

Data Scientist / Quantitative Researcher
Boston,USA

Summary

Data Scientist familiar with gathering, cleaning and organizing data for data exploration, analysis, feature engineering, and modeling. Advanced understanding of statistical, algebraic and other analytical and ML techniques. Highly organized, motivated and diligent with significant experience in data visualization, random forest categorization and time series forecasting.

Overview

17
17
years of professional experience
12
12
years of post-secondary education
7
7
Certifications
2
2
Languages

Work History

Quantitative Researcher / Data Scientist

Apptopia Inc
Boston, MA
03.2022 - 03.2023
  • Led POC time series forecasting project applying a composition of regression models to Apptopia performance metrics and company KPIs to predict quarterly KPI values, e.g., number of subscribers, for apps like Bumble, Spotify, Disney+, Netflix, etc.
  • Helped colleagues, within our FinPod and in Product, hone their quantitative data analysis in projects addressing mobile panel data stability for gathering broad user behavior statistics and assessing our data product statistics
  • Developed polished visualizations to share results of data analyses.
  • Analyzed large datasets utilizing python, Snowflake SQL, and data APIs to identify trends and patterns in customer behaviors.
  • Illuminated our need to go beyond the use of one feature cross, e.g., to measure engagement and highlight that other feature crosses with our performance metrics could provide additional useful information regarding KPIs

Customer Success Engineer

Starburst Data Inc
Boston, MA
08.2020 - 03.2022
  • Engaged with clients to answer questions, submit feature requests, report bugs, discover non-Starburst related issues, and guide their use of Starburst Trino with its various connectors/security options
  • Created knowledge based articles to help fellow engineers, sales reps, and customers streamline their path to production
  • Led an early project to track customer technical needs in order to prioritize the creation of new features in Starburst Trino
  • Followed through with client requests and complaints to resolve problems while prioritizing customer satisfaction and loyalty.

DataOps Engineer

Tamr Government Solutions Inc
Cambridge, MA
09.2019 - 07.2020
  • Utilized DataRobot Platform to develop random forest model marking safe aircraft payload configurations based on ~500 flight logs probing the threshold of catastrophic resonant wing oscillations or flutter in F-16 aircraft
  • Used Tamr mastering ML software tool to train a model to cluster ~1000 military contracts based on their purpose, budget, location, and equipment needs to surface insights about their time and budget risks
  • Interacted with subject matter experts to develop a deeper understanding for data mastering and payload safety certification
  • Read and interpreted blueprints, technical drawings, schematics, and computer-generated reports.
  • Conducted research to test and analyze feasibility, design, operation and performance of equipment, components, and systems.

Fellow

Insight Data Science
Boston, MA
06.2019 - 09.2019
  • Developed and deployed a Flask-based web app on AWS using Facebook Prophet to forecast home buying factors to optimize location choice
  • Trained 131 price forecasters in python for desirable features including home appreciation, local crime level, and wealth of the neighborhood based on Los Angeles zip codes for enhanced city localization
  • Collaborated with other fellows to advance our projects and gain deeper understanding of data science topics.

Visiting Fellow

MIT Kavli Institute for Astrophysics, MLK
Cambridge, MA
09.2017 - 08.2019
  • Created a new stellar chemical abundance, data-driven diagnostic from ~1000 stars to infer the average star formation efficiency of ancient disrupted dwarf galaxies that have been cannibalized by the Milky Way
  • Managed and taught two women international high school students on how to query a NASA web app using curl and python to generate atmospheric chemical abundance (spectral) observations from 800,000 years of Antarctic ice core air bubble samples to detect climate change over human history from light-years away
  • Secured ~ $10,000 in funds to found and direct the MIT Sidewalk Astrogazers, a graduate student outreach team formed to bring astronomy to underserved communities in the Greater Boston Area—reached ~1800 people in two years (https://astrogazers.mit.edu/)
  • Computed complex statistical analyses and interpreted data using python
  • Collaborated with other researchers to write and distribute impactful results of stellar chemical abundance research.

Fisk-Vanderbilt Bridge Fellow

Fisk/Vanderbilt University
Nashville, TN
01.2016 - 08.2017
  • Continued research on developing a stellar chemical abundance ratio distribution diagnostics from semi-analytic stochastic models
  • Mentored three minority astronomy graduate students in the Fisk-Vanderbilt Master's-to-PhD Bridge Program
  • Taught one of my mentees the basics of radio astronomy to give him familiarity with key terms, concepts, and technologies supporting his start to successful radio astronomy research and doctorate in Astronomy
  • Created and developed lesson plans to meet students' academic needs.
  • TA’d and subbed for classes in Physics and Astronomy

Postdoctoral Fellow

Shanghai Astronomical Observatory
Shanghai
11.2013 - 01.2016
  • Applied the expectation-maximization (EM) algorithm with simulated chemical abundance ratio distributions of ~1500 dwarf galaxies of a Milky Way-like galaxy to show that its 12 Gyr dwarf galaxy merger history could be reconstructed from just ~1000 randomly sampled Milky Way Halo stars—stars found in the outskirts of the Galaxy
  • Wrote and published peer-reviewed articles concerning findings and highlighted possible applications for findings
  • Developed unique research in the field of Galactic Genealogy/Archeology
  • Helped to lead our research group meetings on Galaxy Evolution and helped to host international conferences
  • Helped mentor graduate students in our Galaxy group

Graduate Researcher

Columbia University
New York, NY
09.2006 - 09.2013
  • Utilized ~6,000 modeled chemical abundance ratio distributions of dwarf galaxies to predict the number of ancient star observations (~25-30) needed in low-mass dwarf galaxies around the Milky Way before seeing the signature of the elusive r-process—used only 6 low-mass dwarf galaxy stars against ~300 Milky Way Halo stars with a Monte Carlo 2D KS test coupled with the binomial probability function for prediction
  • Theorized a new way of analyzing the kinetic Sunyaev–Zel'dovich effect to gain information about the transverse motions of galaxy clusters—transverse motions being the most difficult to assess in astronomy
  • Identify targets for nuclear star clusters in various large nearby galaxies to probe relationship between these clusters and supermassive black holes at the center of each galaxy
  • Processed radio wave data from the VLA to help identify galaxy cluster environmental effects on the gas content and morphology of disk galaxies in various galaxy groups

Education

Ph.D. - Astronomy

Columbia University
New York, NY
09.2006 - 09.2013

Master of Arts - Astronomy

Wesleyan University
Middletown, CT
09.2004 - 05.2006

Bachelor of Arts - Astrophysics

Williams College
Williamstown, MA
09.1997 - 06.2001

Skills

Data Science, Data Visualization, Random Forest Categorization, Time Series Forecasting, Regression

undefined

Software

Python

C, IDL, MATLAB/Octave, Mathematica, BASH

R

Certification

AWS Cloud Practitioner - CLF-C01 (LA), A CLOUD GURU (17.7 Hrs of Content), VERIFY.ACLOUD.GURU/CC4F5367FFF7

Honors


National Academies of Science Kavli Fellow

Issued by National Academy of Science · Mar 2019

  • Associated with MIT Kavli Institute for Astrophysics and Space Research

Title I Distinguished Graduate Award
U.S. and Massachusetts Department of Education · June 2001

Publications

Refereed Publications

6. r-Process Nucleosynthesis: Connecting Rare-Isotope Beam Facilities with the Cosmos.

Horowitz, C. J.; Arcones, A.; Côté, B.; Dillmann, I.; Nazarewicz, W.; Roederer, I. U.; Schatz, H.; Aprahamian, A.; Atanasov, D.; Bauswein, A.; Bliss, J.; Brodeur, M.; Clark, J. A.; Frebel, A.; Foucart, F.; Hansen, C. J.; Just, O.; Kankainen, A.; McLaughlin, G. C.; Kelly, J. M.; Liddick, S. N.; Lee, D. M.; Lippuner, J.; Martin, D.; Mendoza-Temis, J.; Metzger, B. D.; Mumpower, M. R.; Perdikakis, G.; Pereira, J.; O'Shea, B. W.; Reifarth, R.; Rogers, A. M.; Siegel, D. M.; Spyrou, A.; Surman, R.; Tang, X.; Uesaka, T.; Wang, M., Journal of Physics G: Nuclear and Particle Physics, 46, 8 (2019)

5. Reconstructing The Accretion History Of The Galactic Stellar Halo From Chemical Abundance Ratio Distributions.

Lee, D. M., Johnston, K. V., Sen, B. & Jessop, W., ApJ, 802, 48 (2015)

4. A Mass-Dependent Yield Origin of Neutron-Capture Element Abundance Distributions in Ultra-Faint Dwarfs.

Lee, D. M., Johnston, K. V., Tumlinson, J., Sen, B. & Simon, J. D., ApJ, 774, 103 (2013)

3. Hot and Cold Galactic Gas in the NGC 2563 Galaxy Group.

Rasmussen, J., Bai, X., Mulchaey, J. S., van Gorkom, J. H., Jeltema, T. E., Zabludoff, A. I., Wilcots, E., Martini, P., Lee, D., Roberts, T. P., ApJ, 747, 1, 31 (2012)

2. Measuring Transverse Motions for Nearby Galaxy Clusters.

Hamden, E. T., Simpson, C. M., Johnston, K. V., & Lee, D. M., ApJL, 716, 2 (2010)

1. The Coincidence of Nuclear Star Clusters and Active Galactic Nuclei.

Seth, A., Agüeros, M., Lee, D. & Basu-Zych, A., ApJ, 678, 1 (2008)

Other Publications

5. Understanding the Nature of Chemical Abundance Ratio Distributions in Nearby Stellar Systems.

Lee, Duane M., Ph.D Thesis — Columbia U. (2014)

4. Rooftop Variables: Connecting New York City Astronomers with Public School Teachers.

Hamden, Erika T., Agueros, M., Corrales, L., Hilton, E., Hummels, C., Lee, D., Pereira, M., Saul, D., Zimmerman, N., Dubner, J. (2010)

3. Investigation of Environmental Influence on Galaxian Activity Using KISS and SDSS.

Lee, Duane M., M.A. Thesis — Wesleyan U. (2006)

2. Quantifying Entanglement.

Lee, Duane M., B.A. Honors Thesis — Williams College (2001)

1. The Long KISS Survey for High Redshift Emission-Line Galaxies.

Lee, D. M., KECK Northeast Astronomy Consortium Conference Proceedings (1999)

Timeline

Quantitative Researcher / Data Scientist

Apptopia Inc
03.2022 - 03.2023

AWS Cloud Practitioner - CLF-C01 (LA), A CLOUD GURU (17.7 Hrs of Content), VERIFY.ACLOUD.GURU/CC4F5367FFF7

09-2021

Customer Success Engineer

Starburst Data Inc
08.2020 - 03.2022

Convolutional Neural Networks - Coursera, Credential ID V53JN7BMTHVW

07-2020

Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization - Coursera, Credential ID MAQXKTEWKCSY

06-2020

Structuring Machine Learning Projects - Coursera, Credential ID Z2GVLMSACHLQ

06-2020

Neural Networks and Deep Learning - Coursera, Credential ID BALCQ4FRNRZ7

05-2020

DataRobot Essentials - DataRobot, Credential ID 52110408

10-2019

DataOps Engineer

Tamr Government Solutions Inc
09.2019 - 07.2020

Fellow

Insight Data Science
06.2019 - 09.2019

Visiting Fellow

MIT Kavli Institute for Astrophysics, MLK
09.2017 - 08.2019

Machine Learning - Coursera Course Certificates, Credential ID U2UNMNKDG549

04-2016

Fisk-Vanderbilt Bridge Fellow

Fisk/Vanderbilt University
01.2016 - 08.2017

Postdoctoral Fellow

Shanghai Astronomical Observatory
11.2013 - 01.2016

Ph.D. - Astronomy

Columbia University
09.2006 - 09.2013

Graduate Researcher

Columbia University
09.2006 - 09.2013

Master of Arts - Astronomy

Wesleyan University
09.2004 - 05.2006

Bachelor of Arts - Astrophysics

Williams College
09.1997 - 06.2001
Duane M. LeeData Scientist / Quantitative Researcher