Independent publication // local-first AI // field reports from owned systems
Nodehome
A running publication about local models, private inference, self-hosted agents, weird hardware, research sweeps, and the builders wiring their own AI stack together.
AI is getting physical again. It shows up in terminals, racks, side projects, and ugly little workflows people actually control.
Latest
Daily Sweep - Oct 13, 2026Show HN: SolidCraft - Open-Source Parametric CAD in Rust (With MCP Support)
Agent or tool-use pattern - relevant to local AI workflow automation.
Run fast, private agents on-device with LFM2.5-24B-A2B
Hugging Face blog post - check for new model releases, library updates, or ecosystem shifts.
Intel’s Crescent Island AI Server GPU Gets a Name and Specs: Meet the Intel GPU D-Series
ServeTheHome coverage - server hardware reviews and benchmarks from a trusted source.
New USFP Pluggable Optics Form Factor for Higher Speeds from Cisco and NVIDIA
ServeTheHome coverage - server hardware reviews and benchmarks from a trusted source.
HP launches keyboard PC with Ryzen AI 400 inside, price nearly doubles over 300 series model to $3,000-$4,000
NVIDIA DLSS 5 gets 2X faster with 20% lower VRAM usage, now 10X faster than March demo
Intel Q970/B960 ATX motherboard for Nova Lake appears with up to six PCIe 5.0 slots and an ancient PCI slot
Bus or storage bandwidth topic - relevant to the H12SSL-i's PCIe 4.0 layout.
Field Reports
builds, experiments, notesThree RTX 3090s, One 32B Model: A Pipeline-Parallel Canary
A current field note on why the 3x3090 serving path moved through pipeline parallelism, not tensor parallelism, for the tested 32B AWQ model.
Gemma 4 12B And The Sensory Agent Lane
A public-safe read on Gemma 4 12B as a local sensory preprocessor: useful for seeing, hearing, and structuring observations without turning into an action system.
Hardware
machines, thermals, economicsPower Caps On Three RTX 3090s: Bursts Versus Sustained Load
A measured note on 300W bursty inference, lower caps for sustained runs, and why power-cap sweet spots are workload-specific.
Parallel Agent Serving Is A Hardware Shape Now
A field-report read on 14x RTX 3090 agent serving, EXL3, FP8 KV cache, Aphrodite, and why concurrency is becoming the local hardware metric.