Find Anything — Even When You Forget the Words
Stop hunting through folders and failed keyword searches. Mindweave's semantic search understands what you mean, surfacing the right notes even when you can't remember exactly how you wrote them.
Keyword Search Fails You at the Worst Moment
You wrote a brilliant insight three months ago. Now you need it. You try "productivity," then "focus," then "deep work." Nothing useful comes back. The note is there — you just cannot find it.
Traditional search matches strings, not ideas. One synonym away and your entire knowledge base becomes invisible. The more notes you capture, the harder retrieval gets — the opposite of what a knowledge tool should do.
Meaningful ideas are rarely encoded in predictable keywords. They live in context, in metaphors, in half-formed thoughts. You need a search engine that thinks like you do.
Search by Meaning, Not Memory
Mindweave embeds every note you capture into a high-dimensional vector space using Google Gemini's text-embedding-004 model. When you search, your query is embedded the same way and matched against your notes by semantic similarity — so "staying focused at work" finds notes about Pomodoro timers, flow state, and deep work, even if none of those words appear in your query.
Natural Language Queries
Ask questions in plain English. "What did I learn about negotiation?" returns relevant notes instantly, regardless of original wording.
Gemini-Powered Embeddings
Every note is encoded into 768-dimensional vectors using Google Gemini — one of the most capable embedding models available.
Sub-Second Results
pgvector's HNSW index returns approximate nearest-neighbor results in milliseconds, even across thousands of notes.
Combined Keyword + Semantic
Run full-text keyword search and semantic search side by side. Get the best of both worlds in a single interface.