Yuanjing Shi

Software Engineer at Snowflake & MSCS at UIUC


LinkedIn shingjan
Github shingjan


500 108th Ave NE
Bellevue, WA 98004 US

Work & Experience

Purdue Center for Programming Principles and Software Systems (PurPL)

Research Assistant

2019/05 — 2019/08
  • Worked with Prof. Tiark Rompf on building Flare (OSDI’ 18, SIGMOD’ 18, & Spark + AI Summit 2018) - an accelerator for Apache Spark with native compilation on Scala front-end and support for distributed data processing, data pipelining and streaming.
  • Use Lightweight Modular Staging (LMS) framework to add distributing data processing support with Message Passing Interface (MPI) and outperform Apache Spark by up tp 10 times on standalone and computation-intensive workloads like TPC-H.

Microsoft Research Lab - Asia

Summer Intern

2018/05 — 2018/08
  • Designed and developed GraphView, a middleware and DLL library for key/value databases within one of Microsoft’s core Azure product - CosmosDB, in collaboration with senior researchers, the Intelligent Cloud and Edge Group and the Azure team.
  • Built the testing framework based on YCSB & TPC-C in order to benchmark GraphView and DBx1000 (SIGMOD’ 16), with optimistic concurrency control (OCC) algorithms.


University of Illinois at Urbana-Champaign

2018/08 — 2020/05
Master of Science, Computer Science
  • Advisor: Vikram Adve

  • The Hong Kong Polytechnic University

    2013/08 — 2018/05
    Bachelor of Science (Honours), Computing
  • Advisor: Zili Shao

  • Publications

    An Optimizing Compiler For ONNX Models On Heterogeneous Systems

    Published by IDEALS - University of Illinois, 2020

    Yuanjing Shi

    Leyenda: An Adaptive, Hybrid Sorting Algorithm for Large Scale Data with Limited Memory

    Published by arXiv, 2019

    Yuanjing Shi, Zhaoxing Li

    An Efficient LSM-tree-based SQLite-like Database Engine for Mobile Devices

    Published by IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (IEEE TCAD), 2018

    Zhaoyan Shen, Yuanjing Shi, Zili Shao, Yong Guan

    SQLiteKV: An Efficient LSM-tree-based Lightweight Database Engine for Mobile Devices

    Published by 23rd Asia and South Pacific Design Automation Conference (ASP-DAC), 2018

    Yuanjing Shi, Zhaoyan Shen, Zili Shao


    Programming Lauguages

    Over 20,000 lines
    C/C++, Java, Scala, and Python

    Compiler Infrastructure

    LLVM, TVM, and LMS

    Data Intensive Systems

    Apache Spark and Hadoop MapReduce & HDFS

    Deep Learning Frameworks

    TensorFlow/Keras, PyTorch, ONNX, and TVM