Documentation
Infrastructure（19 articles）
Server hardware, network topology, container orchestration, monitoring, and GPU environment documentation.
Key topics: AMD EPYC 9175F, MikroTik RouterOS, Podman/Quadlet, Ubuntu Server, Prometheus/Grafana, 10GbE Networking, PostgreSQL, LLM Stack Deployment
Latest articles:
2026-04-09
PostgreSQL moved from an on-demand GPU server to a 24/7 Mac Mini. This infrastructure setup retains pgvector and updates connection settings, macOS support and backups.
2026-03-14
Dagster sensors process gateway events independently. NATS JetStream, idempotent storage, and conversation lineage for AI system and data platform development.
2026-03-14
An experiment tracking setup for AI system development using Dagster, MLflow, and MinIO. It covers correlation_id linkage, AirPlay port conflicts, missing PostgreSQL drivers, and an existing database …
LLM Research（43 articles）
Large language model benchmarks, CPU/GPU inference validation, quantization testing, and optimization research.
Key topics: DeepSeek V3.2, Qwen3, Kimi K2.5, GLM-4.7, Llama 4, Hermes, MiniMax, EPYC 9175F inference optimization, GGUF quantization
Latest articles:
2026-10-26
Speed measurements for local LLM deployment. I ran five models on RTX PRO 6000, including Qwen at 256K on one GPU and 1M on two GPUs.
2026-10-15
A local LLM use case: LoRA fine-tuning and NVFP4 quantization with ModelOpt reduced Clef from 54.97 GB to 23.16 GB. Includes mock-form results, five training examples, and an operation video.
2026-10-06
AI integration tests of Jev-Omni decisions on 35 audio, image, Japanese email, and video inputs, with a Clef comparison on the same 25 inputs. Includes additional video tests and saved answers.
Software Tools（10 articles）
Development tools, IDE configurations, MCP integrations, code analysis utilities, and web project implementations.
Key topics: VS Code Server, Zed, Serena MCP, ctree, Dagster, Django, Lightdash, shelpa
Latest articles:
2026-04-09
Stabilizing web research saved to a vault: a UTF-8 panic fix, ONNX model migration, and MCP integration. A record of reusing research in development workflows after LLM integration.
2026-04-09
Nine Rust MCP servers supporting familiar with search, code analysis, and database inspection. Responsibility boundaries, context limits, and asynchronous coordination for LLM integration and AI …
2026-03-30
A Rust MCP server redesigned as filesystem with 14 tools. Trash, undo/redo, and tracking records provide recovery from worker mistakes during AI system development.
Architecture（10 articles）
System architecture designs, distributed pipeline patterns, and migration records.
Key topics: Rust, NATS, Dagster, OpenAI Proxy, SSE Streaming, Go Migration
Latest articles:
2026-05-16
A 55-minute familiar demo generated a Django reservation system. Local infrastructure, model selection, observation, and recovery for AI system development and LLM integration.
2026-04-14
Single-GPU NVFP4 measurements led to replacing a resident llm-jp translator with batches. The record covers LLM resource allocation, throughput, a dual-GPU startup error, and translation configuration …
2026-04-14
A six-page generation run with Claude, Qwen3-Coder-Next 80B and GLM-5.1 exposed a 300-second timeout. The changes cover task splitting, dynamic prompts and execution traces for LLM integration.