Search every word of your Claude.ai history.

Full-text search, 100% local. Find any answer Claude gave you — instantly.

Claude Search ⌥⇧F

transformer architecture

Building a RAG pipeline from scratch

The transformer architecture uses self-attention to process tokens in parallel, which makes it…

May 28, 2026

Fine-tuning vs prompt engineering

…unlike RNNs, the transformer architecture doesn't process sequences step by step, so you can…

April 12, 2026

Explain attention mechanisms

The key insight behind the transformer architecture is that every token can attend to…

March 3, 2026

Search the complete content of every conversation, not just titles. Accent-insensitive and prefix-aware — find any answer in milliseconds.

Every result shows the conversation title, the matching passage with your keyword highlighted, and the date. No clicking around.

Press ⌥⇧G to see your Claude usage over the year, GitHub-style. Spot your most active periods at a glance.

Keyboard shortcuts

Open search from anywhere ⌥⇧F

Open activity heatmap ⌥⇧G

Navigate results ↑↓

Open selected conversation Enter