I help engineering teams get things into production and keep them there — Kubernetes platforms, LLM inference, and cloud foundations on Azure and AWS.
Hands-on engineering for teams running mission-critical workloads.
Three practices, one job: getting systems into production and keeping them there. I work hands-on and embedded with your team — from architecture to day-2 operations, across Kubernetes, AI and cloud.
Cluster design, hardening and day-2 operations you can actually sleep through.
Everything between a working prototype and an LLM system you can run, trust and afford.
Cloud foundations, infrastructure as code and delivery pipelines — built to hand over.
The delivery pipeline I help teams build and operate — automated, observable and GitOps-driven end to end.
Continuous delivery flow
A focused review of your platform, delivery flow and cost — with a clear, prioritized roadmap.
I build alongside your team: clusters, pipelines and guardrails, shipped incrementally.
Runbooks, docs and pairing so your team owns it confidently after I leave.
Tools I work with every day
I’m Stefan Krempel — a platform engineer and the founder of tupevo. For over a decade I’ve designed, built and operated cloud-native infrastructure for startups and enterprises, from a first cluster to platforms serving millions of users.
tupevo is my independent studio. I take on a small number of consulting engagements at a time and channel what I learn into the products we build. You work directly with me — no account managers, no hand-offs.
This very site runs on the stack I recommend: Kubernetes, GitOps with Argo CD, and a privacy-first, cookieless toolchain.
Tell me about your platform and what’s getting in the way. I read every message myself and reply within two business days.
Prefer email? Reach me at [email protected]
LinkedIn