Jiawei Jiang
Software Engineer at MathWorks
Natick, Massachusetts, United States
Summary
🤩
Rockstar🎓
Top SchoolJiawei Jiang is a C++ software engineer at MathWorks with seven years of experience blending physics and computer science to convert real-world problems into high-performance computable models. He holds a master's in computer science from The University of Chicago and a physics bachelor's from Fudan University, giving him a strong foundation in numerical methods and computational thinking. His work spans high-performance computing and computer graphics, with a particular knack for turning creative ideas into visual, performant systems. An active open-source contributor, he has implemented core model execution and data pipelines for urban spatio-temporal libraries like LibCity and added GeomGCN support to PaddleSpatial, reflecting deep experience in graph neural networks and traffic prediction. Based in Natick, MA, he combines production-grade C++ engineering with machine learning research instincts, often focusing on road and traffic representation problems. He enjoys sharing his journey and translating academic research into practical, shippable software.
7 years of coding experience
Bachelor's degree, Physics, graduated, Bachelor's degree, Physics, graduated at Fudan UniversityMaster's degree, Computer Science, Master's degree, Computer Science at The University of Chicago
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (17)
pytorch10
python10
gnn10
machine-learning10
implement10
data-processing10
paddlepaddle10
graph-neural-network10
modeling10
model-driven9
data-pipelines9
deep-learning9
model-driven-development9
data-pipeline9
model-building9
Programming languages (5)
Github contributions (5)
LibCity/Bigscity-LibCity
Dec 2020 - Nov 2022
LibCity: An Open Library for Urban Spatial-temporal Data Mining
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:3 releases, 150 reviews, 124 commits in 1 year 11 months
Contributions summary:Jiawei primarily updated the `abstract_model.py` to include a `get_executor` method, which is essential for model execution, and modified the `utils.py`. These changes suggest the user is involved in the core model structure. The user also contributed significantly to the `trafficdl/data/dataset/traffic_speed_dataset.py` file, indicating a focus on the data processing pipeline and model training. Moreover, contributions related to the `DCRNN` model, `TrafficSpeedPredEvaluator`, and `TrafficSpeedPredExecutor` files demonstrates expertise in the implementation and execution of machine learning models for traffic speed prediction.
estimated-time-of-arrivaletarepresentation-learningdata-miningmap-matching
PaddlePaddle/PaddleSpatial
Apr 2022 - Jul 2022
PaddleSpatial is an open-source spatial-temporal computing tool based on PaddlePaddle.
Role in this project:ML Engineer
ML EngineerContributions:7 commits, 2 PRs, 3 comments in 2 months
Contributions summary:Jiawei contributed to the implementation of a GeomGCN model within the paddlespatial library. Their commits involved adding the model architecture, including layers and forward pass logic, along with supporting utility functions and dataset integration. The user's work is focused on road representation, likely involving spatial-temporal data processing using PaddlePaddle and PGL. The user also updated model names and added references.
information-theorypaddlepaddlespatialspatial-temporaltemporal-computing
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.