Lewis Fishgold
Philadelphia, Pennsylvania, United States
Summary
🤩
Rockstar🎓
Top SchoolLewis Fishgold is a well-rounded machine learning engineer and applied researcher with 11 years of experience building production-ready deep learning systems, particularly for remote sensing and medical imagery. As Senior Machine Learning Engineer at Azavea he co-created and led development of the open-source Raster Vision library, contributing practical data-preprocessing and pipeline improvements used in satellite and aerial imagery workflows. He pairs rigorous academic training from UT Austin with hands-on software engineering dating back to early developer roles, comfortably navigating research, teaching, and product delivery. Based in Philadelphia, he brings a pragmatic focus on model reliability and data quality—evident in contributions like smarter chip filtering, TFRecord tooling, and sensible defaults for RGB handling—that help bridge prototyping and deployment.
11 years of coding experience
17 years of employment as a software developer
Master of Science (M.S.), Computer Science, Master of Science (M.S.), Computer Science at The University of Texas at AustinBachelor of Science (B.S.), Computer Science, Bachelor of Science (B.S.), Computer Science at University of Delaware
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (5)
computer-vision10
machine-learning10
deep-learning10
python10
image-processing10
Programming languages (6)
Github contributions (5)
azavea/raster-vision
Feb 2017 - Nov 2022
An open source library and framework for deep learning on satellite and aerial imagery.
Role in this project:
ML Engineer
ML EngineerContributions:8 releases, 167 reviews, 1566 commits in 5 years 10 months
Contributions summary:Lewis updated logic for testing if a chip is mostly blank in the `make_train_chips.py` module, which suggests involvement in pre-processing training data. They also refactored the train/val split to be based on separate projects in `train.py` and added a TF record make function. The user contributed to changing the default RGB channel order in multiple modules related to image processing and added a command that allows the user to save temporary files in the project.
pytorchgeospatialsemantic-segmentationdeep-learningsatellite
azavea/noaa-hydro-data
Nov 2021 - Jul 2022
NOAA Phase 2 Hydrological Data Processing
Contributions:13 reviews, 67 commits, 17 PRs in 8 months
data-processinghydrologicalnoaaphase-2phase
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.