Experience

TuSimple

• United States
• Software Development
• 400 - 500 Employee

• Software Engineer

  • Jun 2021 - Present



Duke University

• United States
• Higher Education
• 700 & Above Employee

• Teaching Assistant

  • Aug 2020 - Present

  Work as a TA for ECE551 (Programming, Data Structures, and Algorithms in C++). Deliver recitation class each week for students to review C/C++ knowledge. Address students' questions and grade the evaluation assignments Work as a TA for ECE551 (Programming, Data Structures, and Algorithms in C++). Deliver recitation class each week for students to review C/C++ knowledge. Address students' questions and grade the evaluation assignments



NetApp

• United States
• IT Services and IT Consulting
• 700 & Above Employee

• Software Engineer Intern

  • May 2020 - Aug 2020

  1. Improved the performance and scalability of a distributed application on a large cluster for identifying abnormal logs and made the backend production-ready 2. Speeded up the process of filtering and ingesting cases (100MB-1.5TB each) to the Elasticsearch object by performing actions in bulk and by tuning the parameters. Handled 31022 cases within 41 hours. 3. Enhanced the robustness by raising and catching all the errors. The failed request will be recorded after max retries. 4. Added features of pausing and continuing ingestion, keeping track of the status, checking for case updates, etc. 5. Built a new web page based on Flask for users to search the log by files and display results in heat maps. 6. Configured the Elasticsearch environment and encapsulated the whole project in docker 7. Updated progress daily during group standup and reported weekly. Show less



Nanjing University

• China
• Higher Education
• 700 & Above Employee

• Research Assistant

  • Nov 2017 - Apr 2019

  Research assistant at Computational Imaging Technology & Engineering lab at Nanjing University, focusing on image enhancement. Contributions include: - Proposed an image denoising model with 0.31 dB higher PSNR than current DnCNNs model for addictive Gaussian white noise - Designed a noise estimator to attain noise distribution from corrupted images - Proposed a deep residual network with parameter estimation (DRN-PE) to deal with mixed noise of unknown noise types and unknown statistical properties Show less