Jingwei Chew
SDE II at Amazon
Seattle, Washington, United States
Summary
👤
Senior🎓
Top SchoolJingwei Chew is an SDE II based in Seattle with nine years of engineering experience spanning big tech, startups, and ML projects. He builds reliable backend systems—at PayPal he saved roughly 10 developer-weeks per compliance policy change with a config-driven Kafka daemon and kept a high-throughput subsystem stable during prolonged on-call leadership. As a founder/CTO he productionized an e-commerce inventory sync service for multi-tenant use and took it to 15 trial users, and his ML contributions include improving YOLOv4 evaluation and visualization for COCO benchmarks. Comfortable across backend services, mobile workflows, and computer vision prototypes, he pairs pragmatic engineering with a knack for operationalizing research prototypes into production.
9 years of coding experience
3 years of employment as a software developer
Bachelor of Engineering - BE Computer Science, Bachelor of Engineering - BE Computer Science at Nanyang Technological University Singapore
Stackoverflow
326reputation
26kreached
4answers
2questions
Github Skills (15)
yolov410
computer-vision10
pytorch10
coco10
python10
tensorrt9
onnx8
ocr6
text-recognition6
fonts6
api6
amazon-ec26
bitnami6
ppc64le6
tesseract6
Programming languages (10)
Github contributions (5)
Tianxiaomo/pytorch-YOLOv4
May 2020 - May 2020
PyTorch ,ONNX and TensorRT implementation of YOLOv4
Role in this project:
ML Engineer
ML EngineerContributions:6 commits, 2 PRs in 4 days
Contributions summary:Jingwei primarily contributed to the evaluation pipeline of the YOLOv4 model. This included developing a script to evaluate the model's performance on the COCO dataset. They also addressed a bug in category ID matching and refined the test script by adjusting confidence levels. Furthermore, they integrated the evaluation script and visualization capabilities into the main test script, enhancing the model's testing capabilities.
pytorchpytorch-yolov4darknet2onnxyolov4-tinyobject-detection
tehtea/QuickYOLO
Mar 2021 - Apr 2021
A binary-weight-binary-input YOLOv2 implementation based on Larq's QuickNet as the backbone
Contributions:1 release, 42 commits, 4 PRs in 25 days
yolov2backboneobject-detectionlarqyolov5
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.