LLM on Shadowfishhttps://shadowfish.night-tower.net/tags/llm/Recent content in LLM on ShadowfishHugo -- gohugo.ioen-us© 2026 Creeping depths of the deepTue, 04 Aug 2026 18:30:00 +0200DevOps Job Hunting in the Year 3456: Building My Own 70-Person IT Department with Blackjack & LLMshttps://shadowfish.night-tower.net/post/job-hunting-in-the-year-3456/Tue, 04 Aug 2026 18:19:00 +0200https://shadowfish.night-tower.net/post/job-hunting-in-the-year-3456/When recruiters demand a single Junior DevOps engineer who knows 70 tech stacks and 14 cloud providers, the only logical response is to build your own autonomous AI agent team with blackjack and local LLMs.Battle of the Micro-LLMs: Can a 2GB GPU Save You From CPU Agony?https://shadowfish.night-tower.net/post/llm-token-benchmark-showdown/Tue, 28 Jul 2026 10:45:00 +0200https://shadowfish.night-tower.net/post/llm-token-benchmark-showdown/We benchmarked Qwen 2.5, Gemma 2, and Ministral 3 across two homelab machines. Spoiler: Gemma 2 took 140 seconds to load on CPU, giving us time to rethink life choices.