How I Can Help

Main Consulting Topics

AI & Large-Scale Compute

AI is rapidly consuming what used to be traditional HPC topics. The infrastructure challenges — scheduling, memory management, checkpoint/restore, GPU utilization — are converging. I help organizations navigate this intersection:

• AI Memory & Checkpoint/Restore: Integrating memory platforms like MemMachine into AI/ML workflows for efficient context management and fault tolerance
• Container Strategy for AI/HPC: Building, annotating, and distributing optimized container images across heterogeneous compute environments
• Orchestration: Running workloads at scale on Kubernetes, SLURM, or AWS Batch — including bioinformatics and scientific computing pipelines
• Image Curation & Distribution: Managing container images across hardware targets and execution environments. MetaHub was built for exactly this.

Containers & Cloud-Native Infrastructure

Beyond AI and HPC, I am happy to help with container topics in any domain:

• Better build techniques, multi-stage builds, and supply chain security
• Scheduling containers on a single host, a cluster, or across clouds (Kubernetes, SWARM, Batch)
• Troubleshooting container behavior and performance
• Storage consolidation, MinIO on Kubernetes, asset management
• ...and whatever you can think of...

Background

I have worked for automotive R&D companies, HPC vendors (Bull SAS), Playstation Now, Docker (TAM for EMEA), AWS (EC2 Spot Specialist SA and first HPC Developer Advocate), the Max Planck Institute for Human Development, and most recently MemVerge as Principal Architect. A lot of ground covered and a lot of experiences and failures captured. If you have questions about anything — please book a slot and let's have a chat.