Blog
Gedanken zu KI-Systemen, Sprachmodellen und Software, die im echten Betrieb funktioniert.
- Nur auf Englisch verfügbar
What a penetration test costs, according to three public records
The quoted average is $18,300 and traces to a 2023 page that names none of its sources. Sellers who publish put a fixed test at $4,655. The US government paid a median of $39,146 for short engagements. All three are real, and the dataset is public.
security public-records m37 - Nur auf Englisch verfügbar
What changed at the AI frontier: August 2026
The August edition of my monthly reference-doc diff: Google's Imagen shutdown and the migrations that are no longer model-string swaps, OpenAI's GPT-5.6 price cuts recomputed, MiniMax H3's open weights and the licence that excludes the EU, why video-understanding benchmarks are weaker than they look, and the Munich GEMA ruling against Suno.
frontier-delta ai-models deprecations llm - Nur auf Englisch verfügbar
What changed at the AI frontier: July 2026
A scheduled job refreshes my reference docs on AI capabilities every month and commits the diff. The July edition: OpenAI and Google retiring image and video models on short timelines, Voyage's contextualized chunk embeddings, a speech leaderboard reshuffle, Apple's 20B on-device model, and Mistral OCR 4 pricing.
frontier-delta ai-models deprecations llm - Nur auf Englisch verfügbar
Temperature Barely Diversifies an LLM. The Entropy Has to Enter Through the Input.
Sample the same prompt from gemini-2.5-flash ten times and you get about four distinct answers. Raising temperature from 0.2 to 2.0 barely moves that number. What does move it: seeding each call with a different persona, or showing the model its previous answers and asking for a new one. Measured on NoveltyBench, with the honest caveats about where each lever wins.
llm diversity generate-and-select ai-systems -
Was es heißt, einen lauffähigen Demonstrator statt eines Foliensatzes abzugeben
Für einen strukturierten Beschaffer mit einer konkreten Software-Anforderung schlägt ein lauffähiger Demonstrator den Foliensatz. Er lässt beide Seiten die Leistung festlegen und den Beschaffer die Aussagen vor der Bindung prüfen.
prototyping accessibility procurement - Nur auf Englisch verfügbar
Claude Code Shipped Deterministic Agent Orchestration. It's Behind Two Feature Flags.
Claude Code v2.1.147 added a Workflow tool: sandboxed JavaScript that composes subagents with agent(), parallel(), pipeline(), and a real token budget. Here's what it actually does, how to enable it past the dual flag-gate, and what the empirical token cost looks like.
claude-code agent-orchestration ai-systems - Nur auf Englisch verfügbar
The EPD Cost Crisis: Why Small Building Materials Manufacturers Are Getting Locked Out
Environmental Product Declarations cost $17,000-$55,000 per product. For a small manufacturer on 5% margins, that's brutal. Here's where the money goes, why self-service tools don't fix it, and what just changed.
epd building-materials sustainability buy-clean -
KI-Support für Premium-Dienstleister: Das Argument für Erreichbarkeit
Für beratungsintensive Unternehmen geht es bei KI-Support nicht um Kostensenkung. Es geht darum, die Anfragen zu gewinnen, die heute in der Mailbox verloren gehen.
ai customer-support premium-services
Gelegentliche Notizen zu KI in der Praxis
Was sich in der KI-Entwicklung tut und was ich beim Bauen damit lerne. Kein Spam. Jederzeit abbestellbar.