Jon Bratseth
@jonbratseth
CEO http://Vespa.ai Build things and help people.
New Vespa features covered in the June newsletter: - Layered ranking: Rank chunks in documents. - Elementwise bm25 - top, filter_subspaces, and cell_order tensor functions - chunking support in indexing - element-gap: Proximity over chunks - filtering in grouping results -…
Announcing: The RAG Blueprint Build RAG like the world's most successful applications. Start from our open source sample app which contains all you need to do to achieve world-class quality at any scale. Sample app: github.com/vespa-engine/s… Blog post: blog.vespa.ai/the-rag-bluepr…
so much of so-called moral intuition, like that fun wears out and utopia is ultimately boring, is contingently downstream of being permanently imprisoned in rickety rube kludgeberg machine of matryoshka shock collars and dopamine needlepricks for driving a biorobot around a…
The one biological paradox I find really tiresome is that for things to be easy and fun most of the time, you have to intentionally inflict (relatively) stupendous levels of boredom and hardship on yourself.
June @vespaengine newsletter is out! Lots of cool new stuff (e.g. built-in chunking) and educational content (e.g. demo E-commerce apps with new ideas) Check it out and let us know of any feedback: blog.vespa.ai/vespa-newslett…
Correct. And furthermore it's the only humanism.
The one biological paradox I find really tiresome is that for things to be easy and fun most of the time, you have to intentionally inflict (relatively) stupendous levels of boredom and hardship on yourself.
Context engineering x.com/karpathy/statu… + RAG
Introducing layered ranking: RAG meets context engineering.
Training competence getting commodified
It's been quite awe-inspiring to see tiny «community» startups like PrimeIntellect and Arcee jump straight to the frontier of efficiency. The level of the game has been raised very far.
cool to learn @allen_ai are using @vespaengine! and binarized embeddings are amazing for cost/perf, this is a great model
just learned that ai2's scholar qa is powered by mixedbreads embedding model (using binary quantization) and our v1 reranking model. check out the report here: arxiv.org/abs/2504.10861
We'll have time to explain later, but for those who understand, enjoy Vespa 8.530.
A lot more coming around chunking soon. One of those areas where human and LLM needs meaningfully differ.
As most have moved to multi-chunk documents in RAG applications, we thought we should make this easier. Introducing the chunk indexing function:
So much gold in these comments 😂
Introducing VibeCon — the world’s largest vibe coding conference. Make sure you register today: http://127.0.0.1:8080/register
Hard problem indeed. Not everyone wraps google or start from scratch though. Case in point: @perplexity_ai builds on @vespaengine vespa.ai/perplexity/
According to this reuters article, OpenAI is years away from its own web search technology. I think people don't realize how hard it is to build search over the web from scratch, and that's why nearly everyone wraps Google. The challenge: filter trillions of webpages to the N…