Edward Aftandilian

Registered user since Mon 28 Aug 2017

Name:Edward Aftandilian
Bio:

Broadly, I am interested in developer productivity and helping developers write better code.

Currently, I am a researcher at GitHub, where I lead a team focused on improving the quality of completions in GitHub Copilot. GitHub Copilot delivers ML-powered, multi-line code completions in popular IDEs such as VS Code, Visual Studio, and IntelliJ.

Previously, I led the Java Tools and JVM teams at Google. While at Google, I worked on:

  • Error Prone, Google’s static analysis tool for Java
  • Tricorder, Google’s static analysis platform
  • Bazel, Google’s build system
Country:United States
Affiliation:GitHub
Personal website:https://eaftan.github.io
Research interests:Machine learning for code, static analysis, engineering productivity, virtual machines, Java

Contributions

MAPS 2022 Author of Productivity Assessment of Neural Code Completion within the MAPS 2022-track
ICSE 2019 Author of Panel Post-PhD career: What are the tradeoffs between a research career in industry, academia (research track, teaching track), and government? within the Student Mentoring Workshop-track
Author of Analyzing and Repairing Compilation Errors within the Posters-track
Author of Type Migration in Ultra-Large-Scale Codebases within the Technical Track-track
ISMM 2018 Committee Member in Program Committee
SPLASH 2017 Author of Detecting Argument Selection Defects within the OOPSLA-track
PLDI 2017 Committee Member in Program Committee
PLDI 2016 Committee Member in External Review Committee
PLDI 2015 Committee Member in External Review Committee within the Research Papers-track
SPLASH 2011 Author of Asynchronous Assertions within the OOPSLA-track
SPLASH 2010