Jon Bratseth
@jonbratseth
CEO http://Vespa.ai Build things and help people.
First video of the series is up! You’ll learn how to: • Create a Vespa application package • Enable BM25 scoring • Spin up Vespa inside Docker • Feed documents into Vespa from Hugging Face • Run BM25-ranked search queries
I'm a psychologist. Peak math was in high school 😿 Upside: every time a complex problem finally gets decomposed back to familiar territory (Hi, Pythagoras!), I tend to scream like a 4yo at Christmas. There will be quite some math in this lesson. But trust me, you'll get it.
The price of everything on Earth. This chart is all of the natural occurring elements, their occurrence rate in Earth's crust (X-axis) and their price in USD (Y-axis). The chart illustrates three clear price regimes. 1. Yellow band is stuff that is economically priced this is…
Mark your calendars for Tuesday, 6pm CET. This is an event you dont wanna miss! Logan Kilpatrick from Google Deepmind will join me in AI Chitchat! 🚀
Lightning Lessons on The march to Cheat at Search with *Agents* coming in Feb -- Coming Nov 17, Radu Gheorghe of Vespa.ai will share best practices on RAG chunking. Or really beyond RAG chunking :) maven.com/p/e2a6af/beyon…
Best general talk about vectors I've seen. From what a vector is to how HNSW works: youtube.com/watch?v=iIWG3_…
youtube.com
YouTube
Vectors Explained - as non-technical as we could!
I've been looking for how search and RAG can be done on large scale and actual data, and there's just toy examples everywhere I look. Not just some pdfs or a website with everything in context, but actual search, retrieval, ranking, re-ranking, etc. Then I found this goldmine.
Anyone who needs their AI systems to have access to general knowledge will need a web search API. That's probably why Google and Bing are restricting and shutting down theirs now. Fortunately, new alternatives are coming online. Perplexity just launched their web search API…
blog.vespa.ai
How Perplexity beat Google on AI Search with Vespa.ai
Perplexity demonstrates the quality of their search solution and show what it takes to achieve it
Google just made a subtle but massive change Last month, Google quietly removed the num=100 search parameter. This means you can no longer view 100 results at once. The default max is now 10. Why does this matter? - Most LLMs (OpenAI, Perplexity, etc.) rely (directly or…
Filtered vector search is a massively important and overlooked problem for RAG and vector DBs. Very excited to see this new blog post from @vespaengine detailing its implementation of ACORN, along with many clever extensions to deliver huge speedups for search with filters.…
In real vector search systems, performance is dominated by combining it efficiently with filters. Few test this properly. 🧵
Lots of hard problems in web search, but luckily at least the "super fancy db" you need for the index is available for everyone at Vespa.ai.
Why it's hard to build a web index, objectively harder than building a GPT-4.1. Argument: there are just fewer people - literally two (G and M) - who have done it well.
Built on Vespa.ai
Perplexity Search API: Providing direct search results in milliseconds for grounding LLMs and agents with real-time information from the web. This is an effort that began more than two years ago: to build our own search index. So much progress in a short period of time. We look…
Much talk about context rot in timeline. The solution: layered ranking and chunk selection.
In real vector search systems, performance is dominated by combining it efficiently with filters. Few test this properly. 🧵
We just did a podcast about the process of migration (trade-offs included) from #Elasticsearch to @vespaengine With @dainius_jocas and @KevinPetrieTech 🙌 em360tech.com/podcasts/vinte…
People are coming up with so many great uses for layered ranking. Nice to see innovation driven by scaled RAG apps benefiting all kinds of use cases.
Read the full announcement blog here: blog.vespa.ai/introducing-la…
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.
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 -…
United States الاتجاهات
- 1. #AEWDynamite 18.6K posts
- 2. #Survivor49 3,148 posts
- 3. Donovan Mitchell 3,672 posts
- 4. #CMAawards 4,499 posts
- 5. #AEWCollision 6,778 posts
- 6. #cma2025 N/A
- 7. Nick Allen 1,910 posts
- 8. Dubon 3,173 posts
- 9. UConn 7,539 posts
- 10. Okada 11.9K posts
- 11. Cavs 8,273 posts
- 12. Arizona 31.4K posts
- 13. Derik Queen 1,558 posts
- 14. Aaron Holiday N/A
- 15. Bristow N/A
- 16. Lainey Wilson N/A
- 17. Rockets 16K posts
- 18. Mobley 4,223 posts
- 19. FEMA 52.2K posts
- 20. Marquette 2,144 posts
Something went wrong.
Something went wrong.