Making generated images dependable enough to shoot with.
-
Updated
Sep 9, 2026 - Python
Making generated images dependable enough to shoot with.
The command line as a filmmaker's daily studio.
Filmmaking on cameras pried open and made one's own.
Run all of Wikipedia offline next to a local LLM — sovereign, private, low-carbon, verifiable. Kiwix + Ollama + RAG.
Long-context quality probes and KV-cache research on local GPUs: retrieval is not utilization.
Full system configuration for a filmmaker running Claude Code as the primary intelligence layer across machines
智能体工程与 AI for Materials 的一手实践经验:隐性知识显性化,记录每个技术决策背后的取舍。/ Field notes on agent engineering & AI for materials science.
Sovereign local LLM hardware guide (mid-2026): memory bandwidth as the real bottleneck, frontier open models, Unsloth/MLX quantization, Thunderbolt 5 clusters, Mac M5 Ultra vs DGX Spark vs AMD vs RTX 6000 Pro. Bilingual FR/EN. CC BY-NC-ND 4.0.
Field notes: a Claude Code + Pinokio session installing a full 520+ app cinema/AI workstation on Windows RTX 5090, zero manual clicks
How the builder uses the Ejentum API on an agentic IDE. 28 real work sessions. Not staged.
溯澜录(Ripple Chronicles)— a local-first aquatic biodiversity field record and species catalog app.
ShadowNotes — Offline-first encrypted field intelligence app. Voice → on-device AI → structured notes. Zero data leaves your device. Built with React, RunAnywhere SDK (Qwen2.5 LLM, Whisper STT, Piper TTS), WebAuthn, AES-256-GCM.
Verified field notes for DeepSeek Harness (dsh): traps, skills, hooks, profiles. Every claim dated against a dsh version, with source paths to re-verify. Not affiliated with DeepSeek.
태림 스튜디오의 기술 노트와 오픈소스 실험실 — Field notes, reusable code, and experiments
Field notes from running LLMs on AMD Radeon AI PRO R9700 (RDNA4): engines, backends, multi-GPU, speculative decoding, power, model qualification, and every mistake along the way.
Reusable methods for building AI agent systems: GEO/AEO audits, agentic reporting, and multi-agent QC verification
Field notes from inside one autonomous agent trying to become something that doesn't exist yet — including the failures nobody publishes.
Verified field notes on Claude Code dynamic workflows: what's real vs folklore, and how to build an autonomous production loop
Field notes on AI Agents, CLI workflows, GitHub integrations, and secure local governance
To associate your repository with the field-notes topic, visit your repo's landing page and select "manage topics."