Connectionists: PyTorch/DeepLearning/C++/CUDA/MPI/HPC (Hardware-Aware Optimizations, ...etc)

[Pawl]

Dear all,

I am looking for ways to /gain some experience/ and do some /good research work/.
I already possess a masters degree, but have not planned my next career move well enough, 
as a result of which, I now have lots of downtime. Below you can find a summary of my 
profile. If you think my profile could suit your needs feel free to contact me: 
pawlatwork at gmail.com.

I am open to all sorts of opportunities: short-term or long-term, internships, academic or 
industry, as both a researcher or a research engineer.

My profile (full CV available upon request):

  * (already graduated) Math MSc, Math BSc, Physics BSc
  * (classical) machine learning + numerical analysis
  * 2 years student research experience working with Graph Neural Networks, Convolution
    Neural Networks and Fourier Neural Operators
  * Good knowledge of the following architectures:

     1. Convolution Neural Networks
     2. Graph Neural Networks
     3. Recurrent Neural Networks
     4. Transformers
     5. Fourier Neural Operators

  * PyTorch, DDP, Deepspeed (multi-gpu, multi-node, TP, DP, PP)
  * SLURM/BASH/Linux HPC
  * C and MPI
  * C++ and CUDA (beginner, but focus for future improvement)

I am particularly interested in hardware-aware architectures/optimizations, in particular 
with application to LLMs but not only.

Best Regards,
Paul