Aart Bik

Current

• Staff Software Engineer at Google Inc.

Past

• Principal Engineer at Intel Corp.
• Postdoctoral Researcher at Indiana University

Education

• Universiteit Leiden
• Universiteit Utrecht
• Christelijk Lyceum, Gouda

see less...

Recommended

3 people have recommended Aart

Connections

229 connections

Industry

Computer Software

Aart Bik’s Summary

Aart J.C. Bik (a.k.a. Arjan Bik) is an experienced computer scientist with a thorough knowledge of the programming languages C, C++, Fortran, Java, Python, and assembly language for the Intel Architecture. He has worked on a wide variety of architectures and processors, using various development and debugging environments, performance evaluation tools, and operating systems.

Aart Bik’s Specialties:

Compilers, data dependence analysis, high performance computing, large-scale graph computations, parallel and distributed systems, parallelization, SIMD, vectorization.

Aart Bik’s Experience

• Staff Software Engineer

  Google Inc.

  (Public Company; GOOG; Internet industry)

  May 2007 — Present (2 years 7 months)

  Aart took on new challenges and moved to Google Inc. in Mountain View, California, USA, where he is now working as Staff Software Engineer on a distributed system for large-scale graph computations. In his "spare time" he developed chess, checkers, and reversi for the G1 Android phone.

• Principal Engineer

  Intel Corp.

  (Public Company; 10,001 or more employees; INTC; Semiconductors industry)

  January 1998 — May 2007 (9 years 5 months)

  Aart was the lead compiler architect of automatic vectorization in the Intel® C++/Fortran compilers for Windows, Linux, and Mac OS. In March 2002, Aart received the Intel Achievement Award (highest company award) for making the Streaming SIMD Extensions easy to use through automatic vectorization. In April 2005, he was promoted to Principal Engineer. Aart published on vectorization in "The Software Vectorization Handbook" and selected chapters of "The Software Optimization Cookbook".

• Postdoctoral Researcher

  Indiana University

  (Educational Institution; Research industry)

  October 1996 — September 1997 (1 year )

  Aart did research in high-performance compilers for the Java™ programming language under supervision of prof. dr. D.B. Gannon. This research resulted in the implementation of JAVAR (a prototype Java restructuring compiler) and JAVAB (a prototype bytecode parallelization tool).

Aart Bik’s Education

• Universiteit Leiden

  PhD , Computer Science , 1992 — 1996

  During his PhD research under supervision of prof. dr. H.A.G. Wijshoff, Aart designed and implemented the sparse compiler MT1, which is a special kind of source-to-source restructuring compiler that can automatically transform a dense program into a semantically equivalent sparse program.

• Universiteit Utrecht

  MSc , Computer Science , 1987 — 1992

  Propaedeutical exam in computer science on August 29, 1988 (cum laude) and MSc degree in computer science May 25, 1992 (cum laude).

  Activities and Societies:

  A-Eskwadraat

• Christelijk Lyceum, Gouda

  1981 — 1987

  Gymnasium Beta (June 2, 1987).

Additional Information

Aart Bik’s Interests:

Android, astronomy, Commodore 64, computer chess (and other games), GPS, piano, reading, robotics (Lego Mindstorms NXT), solar power

Aart Bik’s Groups:

Member of the Association for Computing Machinery.

•    Alumni van de Universiteit Leiden
•    The Dutch Connection
•    Intel Alumni Association
•    Chessmasters
•    NedIQ
•    ACM Members
•    Alumni of Universiteit Utrecht
•    A-Eskwadraat
•    IT Professionals Netherlands
•    Google Android
•    High Performance & Super Computing
•    C++ Professionals
•    Commodore 64 Enthusiasts
•    Othello
•    Alumni Computer Science, Utrecht University
•    Compiler Experts
•    Mobile Phone Design
•    The Game of Checkers / Draughts / Dames
•    Informatica (Universiteit Leiden)

Aart Bik’s Honors:

C.J. Kok Award (outstanding PhD thesis award), 1996.

Intel Achievement Award, March 2002, for making the Streaming SIMD Extensions easy to use through automatic vectorization.