We are looking for a full time postdoctoral researcher in the Computer Science Department at Purdue to work on a new class of algorithms in numerical linear algebra that are robust to faults and failures in distributed systems and the next generation of exascale machines.

An ideal candidate will have expertise and experience in:
- matrix computations and sparse matrix computations, with some background on the convergence theory of stationary algorithms or Krylov methods
- distributed and large scale implementations of matrix computations in engineering applications, such as linear system solvers, eigenvalue computations, and advanced matrix-based data analytics (e.g. PCA, clustering, etc.)

We will consider exceptional candidates with experience in only one of these areas who wish to develop expertise in the other area.  The postdoctoral advisors are David Gleich & Ananth Grama.

Please send application materials to matrix-postdoc-2017@cs.purdue.edu

Materials should consist of:
- A 1-2 paragraph introduction explaining why you want the position.
- A standard CV with your publications highlighted.
- Two to four people who could provide references.
- (Optional, but encouraged) an annotated CV explaining your specific contribution to 1-3 publications on your CV in 1-2 paragraphs each.
  Highlight the intellectual depth and engineering efforts in your work.

Applications will be screened immediately upon receipt and will be accepted until the position has been filled. The starting date for the postdoc is negotiable, but no later than Fall 2017.  This information can be easily shared through the web-page https://www.cs.purdue.edu/homes/dgleich/matrix-postdoc-2017.html