Three AIs, One Edge Function: How I Structurally Prevent Drift in Parallel AI Work
Three AIs deployed edge functions in parallel without coordinating. The deterministic workflow I built afterwards — and why atomic claiming is the core.
Martin Rau ·
Arbeiten mit KI
Arbeitsweisen, Workflows und Routinen rund um den produktiven Einsatz von KI im Alltag.
Saving Tokens with Claude: 6 Principles That Make Experts Twice as Fast
How I turned my CLAUDE.md from a style guide into a token budget — 6 principles for lower cost, less waiting, and more honest reporting.
Martin Rau ·
Working with AI: one word, two actions — and an hour of debugging
One word, two meanings: I blocked a Git commit, the AI heard deploy. Why a shared vocabulary decides everything when you work with AI.
Redaktion ·
Keeping an AI Knowledge Base Current With RAG: How My AI Stays Up to Date
How I keep an AI knowledge base current with RAG — searchable, maintained by a keyword and re-indexed on its own at night, instead of going stale.
Martin Rau ·
Multi-AI Orchestration Without API Costs: Why Your Existing Subscription Is Enough
Headless agents on your own machine, fed by the subscription you already pay for instead of an API bill — how boostN orchestrates many models.
Martin Rau ·
One Plan Area per Project — With File Attachments the AI Can Read Too
Plans as rich text, files dropped in by drag & drop, and the AI agent reads both. How the new plan area in boostN bundles your project context.
Martin Rau ·
AI Workflows by Keyword: How We Make Recurring Routines Enforceable
A typed keyword triggers a fixed AI routine — and every single step must be committed before the next one appears. Why that's the actual trick.
Martin Rau ·
Controllable Remote Coding Sessions: One Level Above the IDE
The session URL had suddenly vanished from the terminal. How we found it again via a server endpoint — and why it enables working above the IDE.
Martin Rau ·
Optimizing the Token Budget: How a Server-Side HTML Strip Saves 98 %
Raw HTML mockups burn tokens before the AI even understands them. A strip step on the server cut one file from 26,900 to 650 tokens.
Martin Rau ·
When whisper.cpp Swallows the Start of a Sentence — and the Glossary Is to Blame
Our speech recognition kept chopping off half the sentence up front. The culprit wasn't the mic — it was the keyword glossary itself. Proven by an A/B test.
Martin Rau ·
Better speech recognition in boostN-CLI: Audio normalization + a flexible Whisper model
Why dictated text got swallowed, how SoX normalization and a model switcher fixed it — and what is actually happening under the hood.
Martin Rau ·
How we made Speech-to-Text noticeably better with 90 data points and two prompts
How we tuned Whisper-small with 90 data points, two AI prompts and a personalised keyword glossary — copy-paste templates included.
Martin Rau ·
98.64 % uptime? It felt like I was blocked 44 % of the time.
Anthropic's status board shows 98.64 % uptime. The per-day stripes tell a different story — and it's the one that matters for your work.
Martin Rau ·
Usage-based instead of flat-rate: The AI cost turn was predictable — and here is how to absorb it
AI providers are pushing flat rates toward metered billing. Why it had to happen, what it costs and three levers that soften the shift.
Martin Rau ·