Semantic Note Search
Traditional search fails when you can't remember the exact words. You know you wrote something about "the performance problem we discussed" but searching for "performance" returns fifty results. Semantic search understands meaning — ask in natural language and find the right note instantly.
Vector Embeddings
Every note is converted into a mathematical representation of its meaning using Cloudflare Vectorize. Search by concept, not exact text matches.
Natural Language Questions
Type "what did I decide about the database architecture?" and get the specific note where you made that decision — even if it doesn't contain those exact words.
RAG-Powered Answers
Not just finding notes — getting answers. AI retrieves relevant context from your notebook and synthesises a response sourced from your own knowledge.
Relevance Scoring
Results are ranked by semantic similarity. The most relevant notes appear first, not the ones with the most keyword repetitions.
Find Related Notes
Open any note and see semantically related content — other notes about similar topics that you might want to link or cross-reference.
Keyword Search Too
Semantic search works alongside traditional keyword search. Use whichever works for your query — or both together for precise results.
How is this different from Notion's AI search?
Notion's search is still primarily keyword-based with limited AI enhancement. xNotePadAI uses purpose-built vector embeddings via Cloudflare Vectorize, meaning every search query is matched against the semantic meaning of your notes — not just text patterns. And unlike Notion, all your content is encrypted at rest.
Find anything in your notebook. Ask in plain English.
Frequently Asked Questions
How do vector embeddings work?
Each note is converted into a high-dimensional vector (a list of numbers) that represents its meaning. When you search, your query is converted to the same format and compared against all note vectors. Closest matches by meaning appear first.
Do I need a lot of notes for this to be useful?
Semantic search shines with 20+ notes. With fewer notes, you probably remember where everything is. As your notebook grows, the ability to search by meaning becomes increasingly valuable — it scales with your knowledge.
Does semantic search work offline?
No — vector similarity computation happens on Cloudflare Vectorize which requires a network connection. However, standard keyword search works fully offline since your notes are stored locally.
Are my search queries tracked?
No. Zero telemetry, zero analytics, zero logging. We do not track what you search for, how often you search, or what results you click. Your search behaviour is as private as your notes.