Co-op position: LLVM compiler developer
As part of a larger research project within AMD we are looking for a co-op to work on a compiler project that is using LLVM as its backend. The project is studying how explicitly parallel vector programming models (such as Cuda, OpenCL, and ISPC) can be compiled without loss of performance portability. The interested candidate should have experience with LLVM-IR and the LLVM framework. Additionally experience with data-flow analysis, type systems and type inference, and parallel programming is important too.
The position is in Sunnyvale CA, but we could also make it work in Bellevue WA or Austin TX.
Please email me directly benedict.gaster at amd.com with a CV.
- Benedict R. Gaster
Enjoy Berlin and submit a paper to Multiprog 2013: http://multiprog.ac.upc.edu/