Collin Arnett
Machine Learning Engineer at VivSoft
Annandale, Virginia, United States
Summary
👤
Senior🎓
Top SchoolCollin Arnett is a Machine Learning Engineer with eight years of hands-on experience building practical ML and data solutions, currently working at VivSoft while pursuing a BS in Computer Science at George Mason University. He has applied deep learning to real-world problems—from Faster R-CNN sewer-defect detection and YOLOv3 license-plate work at Hitachi Vantara to image classification and production-ready ML pipelines in AWS. Equally comfortable in research-style model work and engineering tasks, he contributes to open-source projects (notably adapting a Keras Faster R-CNN implementation) and explores the intersection of deep learning, Haskell, and DevOps. Collin’s background ranges from teaching K–6 programming to field technical roles, giving him a pragmatic, user-focused approach to shipping robust ML systems.
8 years of coding experience
4 years of employment as a software developer
Associate of Science - AS, Computer Science, Associate of Science - AS, Computer Science at Northern Virginia Community CollegeBachelor of Science - BS, Computer Science, Bachelor of Science - BS, Computer Science at George Mason University
Stackoverflow
37reputation
1kreached
0answers
3questions
Github Skills (13)
object-detection10
computer-vision10
keras10
machine-learning10
faster-rcnn10
tensorflow10
python10
pytorch6
algorithm6
generative-adversarial-network6
google-drive-api6
biginteger6
java6
Programming languages (21)
Github contributions (5)
you359/Keras-FasterRCNN
Sep 2019 - Sep 2019
keras implementation of Faster R-CNN
Role in this project:
ML Engineer
ML EngineerContributions:17 commits, 1 PR, 22 comments in 2 days
Contributions summary:Collin primarily focused on modifying the `transfer/export_imagenet.py` and `keras_frcnn` files to adapt the Faster R-CNN implementation. Their changes included adjustments to the InceptionResnetV2 model, input shapes, and configuration settings for the base network. The commits indicate efforts to fix errors during training, particularly related to the initial epoch and object detection, and adapt the code for testing.
deep-learningr-cnnmachine-learningfastertensorflow
collinarnett/protein_gan
Feb 2020 - Sep 2021
Implementation of "Generative Modeling for Protein Structures" by Namrata Anand and Po-Ssu Huang
Contributions:42 commits, 12 PRs, 39 pushes in 1 year 7 months
anandssuhuanggenerative-modelingbioinformatics
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.