Since LLMs came out, I’ve been amazed at how much faster I can learn. But I wanted one place to truly understand a topic across multiple sources (docs, books, web).
I’m building Kerns to solve this. Spaces are interactive collections that let you:
1. Explore a map from summary to high-detail
2. Switch modalities (text → podcast)
3. Chat with an agent that searches both web + documents
Share and explore community-curated spaces
Examples: Paul Graham’s essays on writing, a comparison of open models, or current news like U.S. visa changes and AI copyright lawsuits.You can create your own space starting from a query and/or sources (pdfs/epubs/htmls).
I'm experimenting with this as a new format for learning and sense-making with AI. Curious if this feels useful to you, please give me feedback.
0 comments
—
Show HN: Kerns – An AI space to understand anything · HackerTrans
I’m building Kerns to solve this. Spaces are interactive collections that let you:
1. Explore a map from summary to high-detail
2. Switch modalities (text → podcast)
3. Chat with an agent that searches both web + documents
Share and explore community-curated spaces
Examples: Paul Graham’s essays on writing, a comparison of open models, or current news like U.S. visa changes and AI copyright lawsuits.You can create your own space starting from a query and/or sources (pdfs/epubs/htmls).
I'm experimenting with this as a new format for learning and sense-making with AI. Curious if this feels useful to you, please give me feedback.