Albert De Poy
Lead Machine Learning Engineer at Macy's
Atlanta, Georgia, United States
Summary
👤
Senior🎓
Top SchoolAlbert De Poy is a Lead Machine Learning Engineer based in Atlanta with 11 years of experience building production ML systems across retail and logistics. He has progressed through data science and senior ML roles at Macy’s and UPS, owning end-to-end pipelines, forecasting and capacity-planning models, and deploying models with cloud and orchestration tools like GCP/AWS and Airflow. A strong back-end developer as well, he contributes to notable open-source projects such as Soot and smali, improving Java/dex tooling and Android compatibility—work that reflects a solid systems-level understanding often missing in pure ML practitioners. His academic background in Operations Research and Industrial Management complements his applied ML skills with rigorous experimental design and optimization thinking. Colleagues describe him as someone who bridges statistical rigor with pragmatic engineering to deliver measurable business impact. He’s equally comfortable tuning a neural network or diving into JVM/dex internals to make production systems more reliable.
11 years of coding experience
7 years of employment as a software developer
Master's degree Operations Research, Master's degree Operations Research at North Carolina State UniversityUniversitat Politècnica de València
English, Spanish, Catalan
Github Skills (15)
javas10
smali10
static-analysis10
soot10
optimisation10
bytecode10
android10
java10
optimization10
dex9
code-optimization9
analyse8
testing8
analyzer8
analyser8
Programming languages (3)
Github contributions (5)
soot-oss/soot
Oct 2014 - Aug 2018
Soot - A Java optimization framework
Role in this project:
Back-end Developer
Back-end DeveloperContributions:56 commits, 15 PRs, 33 comments in 3 years 9 months
Contributions summary:Albert's contributions primarily revolve around modifying the Soot Java optimization framework. They made several changes to the `Scene.java` file, focusing on package exclusion and class inclusion logic. Further work involved adding the ability to exclude single classes. The user also implemented code to ensure stable local ordering and integrated support for reading and processing dex files from various archive formats.
javaoptimizationjava-optimization-frameworkanalysis-frameworkstatic-analysis
JesusFreke/smali
Sep 2016 - Aug 2018
smali/baksmali
Role in this project:Back-end Developer
Back-end DeveloperContributions:9 commits, 5 PRs, 28 comments in 1 year 10 months
Contributions summary:Albert primarily contributed to the `smali` project by addressing resource management and compatibility issues. They fixed resource leaks by ensuring files were properly closed after use, particularly in `smali` and `dexlib2`. Moreover, the user updated the codebase to support Android 8.1 (OAT 131) and fixed instructions related to method handle operations, enhancing compatibility and performance. Furthermore, the user updated tests to reflect code changes.
smali
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.