senn-techTopics
Blog topics
The posts grouped by topic: language models on our own hardware, AI in production, servers and networking, storage and backup, security and data protection, put-into-context AI news, smart home.
33 postsLocal AI and GPU hardwarePosts on running language models in-house: GPUs, mini workstations and inference servers, with measurements for throughput, power draw and cost per token.58 postsAI in productionRAG, agents and automation in the company: which setups hold up in operation, what needs maintenance, and where the models stop working.47 postsServers, Proxmox and networkVirtualisation, clustering, DNS and networking run in-house: Proxmox, high availability, resolvers, updates and the outages that come with them.9 postsStorage, backup and replicationThe storage journey from Ceph through VitaStor to DRBD and LINSTOR, plus ZFS fileservers, backup paths and what storage costs to run.45 postsSecurity and data protectionMail gateway, virus scanning, blocklists and perimeter hardening, plus data protection and EU AI Act practice for AI systems in operation.55 postsAI explained: models, market, newsNew models, releases and vendor news, judged by what they change in day-to-day operation.7 postsSmart home and building automationHome Assistant, Loxone and radio-based devices in residential builds: wiring, switches, heating and integration with your own systems.