Siwei “Robert” Li
Education
Georgia Institute of Technology, Atlanta, GA
2016 - 2020
Experience
Google, San Francisco, CA
June 2020 - Present
Building and improving mission-critical internal data pipeline systems.
Google, San Francisco, CA
Summer 2019
Built an RPC service and a web application adding explainability to a machine learning anomaly detection model in order to generate intuitive alerts for abnormal enterprise Google Doc access.
Georgia Institute of Technology, Atlanta, GA
Fall 2017 - Spring 2020
Advisor: Prof. Duen Horng (Polo) Chau
Member of the Polo Club of Data Science, where I innovate at the intersection of data science and human-computer interaction (HCI) to synthesize scalable, interactive, and interpretable tools that amplify human’s ability to understand and interact with big data.
Panasonic Automotive Systems, Peachtree City, GA
Summer 2017
Developed an extensible analytics tool in Java/Electron/React/d3 that enables engineers to create automated sequences that generate various analysis reports; estimated to reduce 30% of testing time.
Publications
Argo Lite: Open-Source Interactive Graph Exploration and Visualization in Browsers
CIKM'20
29th ACM International Conference on Information and Knowledge Management (CIKM'20), October 19–23, 2020, Virtual Event, Ireland.
Track: Resource
MLsploit: A Framework for Interactive Experimentation with Adversarial Machine Learning Research
KDD'19
ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD). Aug 4, 2019. Anchorage, Alaska USA.
Project Showcase, Oral presentation
Mixed Reality for Learning Programming
IDC'19
18th ACM Conference on Interaction Design and Children, Jun. 2019.
Poster, Honorable Mention
Shield: Fast, Practical Defense and Vaccination for Deep Learning using JPEG Compression
KDD'18
ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD). Aug 19, 2018. London, UK.
Track: Applied Data Science, Audience Appreciation Award Runner Up
Compression to the Rescue: Defending from Adversarial Attacks Across Modalities
KDD'18
ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD). Aug 19, 2018. London, UK.
Project Showcase
Selected Projects
Intel MLsploit, Adversarial Machine Learning Research Platform
Fall 2018 - Spring 2019
NIH mHealth Visual Discovery Dashboard, Part of NIH's MD2K (Mobile Sensor Data-to-Knowledge) Initiative
Spring - Fall 2018
Argo Graph, Large-scale Graph Visualization Tool
Fall 2017 - Spring 2020
Selected Personal and Open Source Projects
NSF Smart Meter Data Portal: Searchable Github-based data-sharing platform for energy researchers
Spring 2018
HackGT/SponsorshipPortal: Intelligent portal for hackathon sponsors to search resumes of participants
Fall 2017 - Fall 2018
Beginners-Cordova-Tutorial: Website teaching Apache Cordova development to beginners, written in Sphinx
Spring 2017
V-Plate: Voice-controlled food delivery robot built with RaspberryPi 3 with an Android client
Spring 2017
On-campus Activities
VIP Project: Health Informatics on FHIR, Atlanta, GA
Fall 2018 - Fall 2019
Develop FHIR applications to improve Emergency Room and Triage efficiency. Currently working with Grady Memorial Hospital, the fifth-largest public hospital in the US, with one of the busiest ER in the country.
HackGT Organizers Team, Atlanta, GA
Spring 2017 - Fall 2019
Hold hackathons and workshops; make open source projects for event organizers; mentor beginners and high school students.
Awards/Scholarships
Outstanding Undergraduate Research Award, Georgia Institute of Technology College of Computing
Spring 2020
President's Undergraduate Research Award (PURA), Georgia Institute of Technology
Fall 2019
Selected Skills
Languages: Java, Python, JavaScript (ES6+), TypeScript, Kotlin, Swift, C++, SQL, Bash, PHP, Go
Web: React, Redux, Node/Express, Electron, Django, Flask, CodeIgniter, HTML, CSS, JQuery
Mobile: Android, Cordova, iOS/ARKit
Data Science: Machine Learning (scikit-Learn, Tensorflow, numpy, pandas), Data Visualization (D3, matplotlib)
Miscellaneous: Infrastructure (Docker, Kubernetes, Vagrant), Build Systems (Gradle, Webpack, Travis CI)