Frank Vitetta | Private AI and data readiness consultant

# frankvitetta.net > AI implementation consultant in London. I help teams get their data ready for AI and run AI privately. I also explain big data in plain language. This is the personal site of Frank Vitetta, an AI implementation consultant based in London. Every article cites its sources at the end. Contact is through the form at https://frankvitetta.net/contact/. ## Field notes - [Open table formats explained: Iceberg, Delta, Hudi](https://frankvitetta.net/field-notes/open-table-formats-iceberg-delta-hudi/): What a lakehouse is. How Apache Iceberg, Delta Lake and Apache Hudi turn files into tables. Where each format stands in October 2026. - [Prompt length vs brand mentions: 8,555 ChatGPT answers](https://frankvitetta.net/field-notes/prompt-length-brand-mentions-chatgpt/): 8,555 ChatGPT responses by prompt length. Mention rate peaks at 35 to 69 characters and is 6.5% at 120 and over. One weak bucket rests on 40 responses. - [When data is not big: DuckDB and single-machine analytics](https://frankvitetta.net/field-notes/when-data-is-not-big-duckdb/): Single-machine engines such as DuckDB question whether analytics needs a distributed cluster. The evidence on both sides, with sources. - [Meta Muse and AI agents: an 11-point website checklist, checked](https://frankvitetta.net/field-notes/meta-muse-ai-agents-checklist/): A viral 11-point list says how to get recommended by Meta Muse. Each point checked against what Meta, OpenAI, W3C and Google document. - [Data quality before AI: dimensions and standards](https://frankvitetta.net/field-notes/data-quality-before-ai/): Retrieval and analytics systems inherit the flaws in their data. The six quality dimensions and what ISO/IEC 25012, ISO 8000 and DAMA UK define. ## Pages - [Home](https://frankvitetta.net/): AI implementation consultant in London. I help teams get their data ready for AI and run AI privately. I also explain big data in plain language. - [Code Scout: a free local AI coding agent](https:/