Resume

Updated in May, 2018


PDF version resume (Updated on in April 2017):
Resume of Quanlai Li


Quanlai Li

E-Mail | LinkedIn | GitHub | Personal Site

 

Summary

Technical Skills:

  • Programming Languages: Proficient in Java and C++, Familiar with Python, JavaScript, PHP, Lua, Scala
  • Computer Science: Machine Learning, Software Engineering, Information Retrieval, Image Processing, Data Structure
  • Others: Linux shell, Windows PowerShell, git, LaTeX, SQL

Soft Skills: Project Management, Software Requirement Analysis, Engineering Leadership, Team Building

Experience

Uber - Software Engineer, Jul. 2017 – Now

  • Building a data workflow management tool

Club Factory - Software Engineer, Apr. 2016 – Jun. 2016

  • Built a software testing automation framework and a version control infrastructure
  • Deployed a Selenium standalone server and a Jenkins server that automatically runs the test cases triggered by commits

Insigma - Research Engineer, Nov. 2014 – Apr. 2016

  • Analyzed issue reports from open-source community and labeled them as bug, feature, documentation update etc.
  • Addressed the problem that 39% of issues are misclassified as bugs which introduces a bias in bug prediction models
  • Estimated the impact of bug misclassification and recommended manual data validation to filter out non-bugs

Education

UC Berkeley – Master of Engineering in Computer Science,  2016 – 2017

  • Concentration: Data Science and Systems
  • Capstone Project: Scaling Up Deep Learning on Clusters
  • Technical Courses: Machine Learning, Database

Zhejiang University – Bachelor of Engineering in Software Engineering, 2012 – 2016

  • Higher-Level GPA:3.95, Grade Point Rank: 1/92
  • Awards: National Scholarship (Top 1.8%)

North Carolina State University - Exchange in Global Leadership Institute, 2016

  • Award: Ye-Li International Exchange Full Scholarship (2 Scholars Annually)

Singapore Management University - Exchange in School of Information, 2015

  • Award: Temasek Asian Young Leader Scholar

Publication

Towards a Trusted Social Network with Blockchain Technology (SCFAB 2018), Mar. 2018

Detecting Similar Repositories on GitHub (IEEE SANER 2017), Nov. 2016

  • Invented a GitHub repository recommendation algorithm based on two heuristics leveraging data not considered before
  • Developed the corresponding system, RepoPal based on a large amount of data mined from GitHub
  • Demonstrated that RepoPal outperforms CLAN (state-of-the-art) in precision by 20% and confidence by 41%
  • Analyzed other advantages of RepoPal, including non-language-specific and less computation cost

Project

Scaling Up Deep Learning on Clusters - Research Engineer, Aug. 2016 – Present

  • Extending state-of-the-art machine learning library, BIDMach on clusters of cloud machines with GPUs
  • Improved BIDMach distributed communication system that allows clusters to run model-parallel algorithms
  • Implementing and benchmarking machine learning algorithms (e.g. Random Forest and K-Means) on BIDMach
  • Integrating BIDMach to OpenChai’s mobile-centric hardware to provide a more affordable machine learning solution

Read More…

Interactive Digital Montage – Software Engineer, Feb. 2015 – Jul. 2015

  • Constructed an image fusing software system based on algorithms including Graph Cut and Gradient Domain Fusion
  • Optimized the parameters of data term and smooth term to lower cost function, providing the best result
  • Explored a new feature and the corresponding cost function that can reduce interaction by 70% and running time by 60%
  • Refined the approach by using a local Gaussian mask, reduced artifacts of fused images when the smooth area is split

Read More…

AppNowIndependent Developer, Jul. 2015 – Aug. 2015

  • Designed a mobile App search engine that solves the mismatch between its product name and underlying function
  • Utilized information retrieval algorithms (e.g. Page Rank and tf-idf) to efficiently and effectively search through dataset
  • Introduced new resources of data including description and reviews of an App, and assigned different weights to them
  • Demonstrated that AppNow outperforms than Google Play search engine in precision and recall based on given dataset

Read More…