GPU Infrastructure at tech events
35 write-ups · 18 events · 2024–2026
I collect sessions on GPU infrastructure here: sharing and scheduling GPUs, multi-tenant clusters, NVIDIA tooling, and the hardware underneath AI workloads. Themes that repeat across KubeCon, Red Hat Summit, RISC-V and AI meetups are how to keep expensive accelerators busy, how to share them safely between teams, and what the platform around them looks like.
2026
22 Sept 2026 · AI Tech Summit "Filip Avramchev" 2026 · Skopje
AI Tech Summit Skopje 2026: From AI Demo to ProductionMy keynote at AI Tech Summit Filip Avramchev 2026 in Skopje, the sessions I photographed, and a practical checklist for taking AI from demo to production.
29 Jul 2026 · KubeCon + CloudNativeCon Japan 2026 · Yokohama
KubeCon Japan 2026 Keynotes: AI Platforms at ScaleThe KubeCon Japan 2026 day 1 keynotes from Yokohama: SoftBank's infinite agents, Fujitsu's GPU-centric Kubernetes, PFN multi-tenancy and Hyundai's Argo CD.
29 Jul 2026 · KubeCon + CloudNativeCon Japan 2026 · Yokohama
KubeCon Japan 2026: the CNCF Briefing and Lightning TalksFrom the KubeCon Japan 2026 press briefing: Japan's 950K cloud native developers, HAMi and Confidential Containers in incubation, and the IOWN partnership.
29 Jul 2026 · KubeCon + CloudNativeCon Japan 2026 · Yokohama
Open Source Building Blocks for AI Platforms (KubeCon Japan 2026)From KubeCon Japan 2026: the AI conformance framework for portability, a CNCF+in-house research platform, and dynamic resource composition in CNCF sandbox.
29 Jul 2026 · KubeCon + CloudNativeCon Japan 2026 · Yokohama
Photonic Networks: the Standout Tech at KubeCon Japan 2026From KubeCon Japan 2026: photonic networks could reshape the AI data center by moving data optically — Fujitsu leads this work in Japan.
29 Jul 2026 · KubeCon + CloudNativeCon Japan 2026 · Yokohama
Tuning Kubernetes for AI: the Real Trade-offs from KubeCon Japan 2026From KubeCon Japan 2026: why AI on Kubernetes needs tuning, not just more GPUs — LLMD caching/routing and the real GPU, memory and electricity cost.
29 Jul 2026 · KubeCon + CloudNativeCon Japan 2026 · Yokohama
Post KubeCon Japan 2026: Reflections on AI, Cloud Native, and the Developer JourneyStraight from KubeCon Japan 2026: the shift from CPUs to GPUs, exploding AI models, and the new accelerator landscape — what it means for developers.
28 Jul 2026 · Cloud Native Telecom Meetup Japan 2026 · Tokyo
Cloud Native Telecom Meetup Japan 2026 at NTT DOCOMO Open Lab Odaiba: My RecapCloud Native Telecom Meetup Japan 2026: docomo agentic AIOps, a 300+ cluster vRAN upgrade, KDDI on CNFs, DRA/DRANET, LF Networking and AI energy metering.
25 Jun 2026 · PlatformCon 2026
PlatformCon 2026: Multi-Tenant GPUs on OpenShift AILessons orchestrating multi-tenant GPUs on OpenShift AI with NVIDIA KAI: GPU sharing, workload isolation, scheduling efficiency, and cost control.
8 Jun 2026 · RISC-V Summit Europe 2026 · Bologna
EPIC Semi's RISC-V AI Server Runs UbuntuAt RISC-V Summit Europe 2026 I met Chloe Jian Ma of EPIC Semi to discuss the first RISC-V AI server — 48 cores, 16 AI cores, and Ubuntu booting live.
8 Jun 2026 · RISC-V Summit Europe 2026 · Bologna
InspireSemi & NextSilicon: RISC-V HPCA show-floor look at RISC-V Summit Europe 2026: InspireSemi's RISC-V supercomputing accelerators and NextSilicon's high-performance compute for HPC and AI.
8 Jun 2026 · RISC-V Summit Europe 2026 · Bologna
RISC-V Summit Europe 2026: Hardware Innovation from BolognaMy experience at RISC-V Summit Europe 2026 in Bologna — from Tenstorrent Blackhole to Monte Cimone v3 HPC, ESWIN server CPUs, and open-silicon research.
8 Jun 2026 · RISC-V Summit Europe 2026 · Bologna
RISC-V: From Dev Boards to AI ServersA floor tour at RISC-V Summit Europe 2026 showing how RISC-V now scales from small embedded dev boards to Tenstorrent AI compute and Scaleway servers.
2 Jun 2026 · AI on the Amstel (June 2026): frontier models panel · Amsterdam
AI on the Amstel: DeepMind, NVIDIA & Mistral PanelRecap of the June 2026 AI on the Amstel meetup at VU Amsterdam — a panel with Google DeepMind, NVIDIA, and Mistral on how frontier models are built.
June 2026 · Red Hat Tech Day Netherlands 2026
Red Hat AI Model-as-a-Service with llm-dHow Red Hat's llm-d transforms LLM inference into a composable Kubernetes-native architecture: disaggregated serving, smart autoscaling, and MaaS.
June 2026 · Red Hat Tech Day Netherlands 2026
vLLM Inference Optimizations on Red Hat OpenShift AIDeep dive into vLLM inference optimizations: KV cache, continuous batching, quantization, and distributed inference with Tensor Parallelism.
19 May 2026 · AI Security Night Amsterdam · Amsterdam
AI Security Night Amsterdam: Breaching LLM-PoweredBrian Vermeer from Snyk presented 'Breaching LLM-Powered Applications' at AI Security Night Amsterdam, covering prompt injection, data privacy, agentic AI.
12 May 2026 · Red Hat Summit 2026 · Atlanta
Summit 2026 Keynote: Metal to Agents and TokensNotes from the Red Hat Summit 2026 day one keynote: open models, token economics, the metal to agents stack and an NVIDIA conversation on agent governance.
11 May 2026 · Red Hat Summit 2026 · Atlanta
Building Digital Sovereign AI: Red Hat and MetaX at RedLi Ming Tsai and Jiaju Zhang present China's sovereign AI stack at Red Hat Summit 2026 — MetaX GPUs, Qwen/DeepSeek powering 30% of global tokens.
11 May 2026 · Red Hat Summit 2026 · Atlanta
GPUs Take Flight: Multi-Tenant Platform Engineering atI presented safety-first multi-tenant GPU platform engineering with NVIDIA and Red Hat OpenShift AI at Red Hat Summit 2026 Discovery Theater in Atlanta.
11 May 2026 · Red Hat Summit 2026 · Atlanta
Jessie Lacome on Lenovo's ThinkSystem SR675 V3 AI ServerJessie Lacome walks through Lenovo's ThinkSystem SR675 V3, an AMD EPYC server for OpenShift Virtualization and AI headroom.
11 May 2026 · Red Hat Summit 2026 · Atlanta
llm-d at Red Hat Summit 2026: KV-Cache Aware Routing for vLLMRed Hat presented llm-d at Summit 2026: cache-aware load balancing for vLLM. Cold 4.3s vs warm 0.6s (7x faster), $0.30 vs $3.00 per 1M tokens with caching.
11 May 2026 · Red Hat Summit 2026 · Atlanta
Red Hat NVIDIA AI Factory: OpenShift Sandboxed ContainersInside the Red Hat and NVIDIA joint session at Summit 2026 — OpenShift sandboxed containers, confidential compute, and AI Factory workload projection.
11 May 2026 · Red Hat Summit 2026 · Atlanta
Red Hat Summit 2026: GPU Platform Engineering TalkLuca Berton presents 'GPUs take flight: Safety-first multi-tenant Platform Engineering with NVIDIA and Red Hat OpenShift AI' — a lightning talk at Red Hat.
7 May 2026 · DevWorld Conference 2026 · Amsterdam
DevWorld 2026: The Next Generation of AI Is Powered byA DevWorld 2026 keynote argued that the next generation of AI depends on infrastructure, not models — six pillars from latency to 100K req/s scale.
22 Apr 2026 · Platform Engineering Meetup NL (inaugural) · Amsterdam
Platform Engineering Meetup NL: Multi-TenantI am presenting 'Lessons Learned Orchestrating Multi-Tenant GPUs on OpenShift AI with NVIDIA KAI' at the first Platform Engineering Meetup NL in Amsterdam.
24 Mar 2026 · KubeCon + CloudNativeCon Europe 2026 · Amsterdam
Speaking at KubeCon Europe 2026Luca Berton presents 'Lessons Learned Orchestrating Multi-Tenant GPUs on OpenShift AI with NVIDIA KAI' at KubeCon + CloudNativeCon Europe 2026 in London.
23 Mar 2026 · KubeCon + CloudNativeCon Europe 2026 · Amsterdam
KubeCon Europe 2026 Recap: My Week in PhotosKubeCon Europe 2026 in photos: co-located days, two Kubernetes Recipes book signings, my multi-tenant GPU talk on OpenShift AI, and the keynote hall.
2025
8 Oct 2025 · World Summit AI 2025 · Amsterdam
World Summit AI 2025 Amsterdam: Data Centres and AgentsWorld Summit AI 2025 at Taets, Amsterdam: Karen Hao on the Empire of AI, $3–8T data-centre capex, Groq on sovereign AI, quantum-safe crypto and agents.
28 Aug 2025 · AI_dev Europe 2025: Open Source GenAI & ML Summit · Amsterdam
AI_dev Europe 2025 Amsterdam: CERN, Cerebras and LanceDBAI_dev Europe 2025 at RAI Amsterdam: CERN's MLOps platform, five Cerebras inference lessons, LanceDB's multimodal lakehouse and Neo4j graph agents.
5 Jun 2025 · AI Salon Amsterdam (8th edition) · Amsterdam
AI Salon Amsterdam June 2025: AI Factories and PitchesAI Salon Amsterdam, 5 June 2025 at The Flow: Nebius on the US-EU AI funding gap, supercomputers versus AI factories, and a round of one-minute demo pitches.
10 Apr 2025 · KubeCon EU Recap and special guests from Nutanix and AWS (Dutch Cloud Native & AI Community Group) · Hoofddorp
Dutch Cloud Native Recap: Nutanix NKP and AWS TritonDutch Cloud Native & AI meetup, Hoofddorp, 10 April 2025: air-gapped Kubernetes bootstrapping with Nutanix NKP, then Triton on EKS with Karpenter from AWS.
3 Apr 2025 · KubeCon + CloudNativeCon Europe 2025 · London
Benchmarking GPU Workloads on Kubernetes with TritonA KubeCon London 2025 talk on benchmarking AI and GPU workloads in Kubernetes: Triton, GenAI-Perf, fmperf and a time-slicing versus MPS comparison.
1 Apr 2025 · KubeCon + CloudNativeCon Europe 2025 · London
KubeCon London 2025: DRA GPU Sharing and CERN KubeflowFrom the AI co-located day at KubeCon London 2025: NVIDIA DRA driver examples for GPU sharing and MIG, and how CERN runs ML challenges on Kubeflow.
2024
25 Oct 2024 · Ubuntu Summit 2024 · The Hague
Ubuntu Summit 2024 in The Hague: AI, RISC-V and SnapsDay 2 of Ubuntu Summit 2024 at the World Forum, The Hague: Intel's OPEA, Google's TCPX, KDE Plasma as snaps, Penpot's Rust pivot and Ubuntu Kylin AI OS.