Open to Staff roles in AI infrastructure
Platform Engineer
for AI Infrastructure
I lead the infrastructure team in a regulated healthcare environment, where I own our production LLM gateway. On my own time I am building a spot-GPU inference lab on EKS to measure what self-hosted LLM serving actually costs.
Amsterdam, NetherlandsmyTomorrows
What I am working on
Production LLM gateway in healthcare
I own the LLM gateway our production AI services run through, with EU-resident model endpoints and a cross-provider fallback, in a regulated healthcare environment.
See the workDebugging GPU scheduling
The LimitRange that requested a GPU for every pod: the root cause, why GitOps could not see it, and a two-minute repro on kind.
Read the postGPU inference lab on EKS
Spot A10G nodes, vLLM on KServe and a signed supply chain, rebuilt from empty state each session. The first GPU benchmark is still to come; I publish measurements, not targets.
Visit the labImpact
- €900K+Annual savings across two platform programmes
- 200+Microservices on the platform I led (PVH)
- 20+Engineers mentored
- 3Clouds run in production (AWS, Azure, GCP)