Hello! I am a doctoral candidate at Vrije Universiteit Amsterdam, member of the VUHPDC group, and also involved with the Computer Networks group at the Paderborn University.

My interests lie at the intersection of HPC/AI, networking and programming languages, currently working on in-network computing.

You can check my publications on Google Scholar and codes on Github

Some technologies i am interested in (in no particular order):

• GPUs/Accelerators, Programmable switches, DPUs
• C/C++, LLVM, P4, CUDA, MPI, DPDK, RDMA

Some slides:

• [PRESENTATION] Supercomputing'24 - NetCL: A Unified Programming Framework for In-network Computing
• [PRESENTATION] Hotnets'21 - Don't you worry 'bout a packet: Unified Programming for In-network Computing
• [PRESENTATION] Netys'21 - Fault-tolerant termination detection with Safra’s algorithm
• [PRESENTATION] Memory Access Monitoring of GPU kernels (MSc thesis)

Small cluster with programmable networking equipment that i maintain and run most of my experiments on:

    
     (click on the image to enlarge)

I am generally reachable at g.karlos AT vu.nl

You, the reader, are particularly encouraged to contact me if you:

• are a BSc/MSc student looking for a project and (a) found anything of the above interesting or (b) have some cool HPC/AI workload you think can be accelerated inside the network
• are a practitioner in the area of programmable dataplanes and think there is room for collaboration. I am particularly looking into getting more involved with DPUs and DOCA

PS. This website will be properly populated soon™!