Ai

Introduction Nine namespaces in my cluster send work to a local language model. The SDR research stack tags transcriptions, the politics dashboard summarises feeds, the congressional-trade tracker prepares daily summaries, and several agents submit work throughout the day. They all call one stable hostname on port 11434. Behind it are two very different Ollama deployments. The primary is an RTX 5090 in a desktop tower that I deliberately power down at intervals. The fallback is an NVIDIA GB10 Spark board, where CPU and GPU share memory and GPU allocations count against the pod’s memory limit. The repository also contains a CPU-only manifest, although the active Kustomization excludes it. The Spark is the backend that remains when the tower is dark. ...

Introduction Consider an agent running in a cluster or on a server that needs to read the news. The customary response is to provide an HTTP client and permit it to fetch URLs, perhaps through an egress proxy intended to impose a measure of network hygiene. The agent in question is an in-cluster assistant with material access to its environment. It can receive a projected ServiceAccount token, whose effective authority derives from Kubernetes RBAC; it can control home automation; and it can reach several important internal services. Giving a system with that degree of reach a general-purpose fetch(url) tool creates a direct security problem: every page it reads becomes a potential source of adversarial instruction, and every host on the internet becomes a potential destination for whatever information the agent can assemble. ...

Introduction Members of the United States Congress must disclose many securities transactions, and the resulting records are public. As a dataset, however, they are close to unusable: the House publishes a ZIP archive whose underlying filings are frequently scanned images; the Senate uses a separate system with its own irregularities; tickers are inconsistent or absent; and neither chamber provides an interface resembling a durable public API. Several commercial services solve this and sell the result. I wanted to see how much of it I could do myself, in-cluster, from the primary sources. ...

Introduction I use Claude Code for substantial one-off analysis, and its output is frequently a single self-contained HTML file. Examples include a flight-log dashboard, an interactive trainer derived from an audiobook transcript (described in Building Interactive Trainers From My Audiobook Library), a pedigree chart, and a client map. Each began either as a Live Artifact or as a file in /tmp that I opened with file://, inspected once, and subsequently lost. ...

Building Interactive Trainers From My Audiobook Library I listen to a considerable number of business and leadership audiobooks, yet retain very little of their content a month later. Listening is passive; useful frameworks blur together with time; and “I read that one” readily becomes “I think I read that one.” I wanted the retention without replaying seven hours merely to recover the one idea I needed. I therefore built a compact production process that converts an audiobook into an interactive trainer, in which each concept and example becomes a card supporting three modes: Browse, Flashcards, and Quiz. Each book receives its own self-hosted site on my internal network; during a single afternoon, I applied the process to seven titles. ...