Carlos Delgado
@carlosdelest
Software Architect and Developer. Sometimes I even pass the Turing test.
Bạn có thể thích
Elasticsearch is under-appreciated and not particularly well understood. It’s worth getting familiar with if you’re dealing with a lot of documents, particularly for agentic search workflows with LLMs via MCP tools.
Elastic v9 is actually really awesome, you should give it another shot. You can do Algolia/Typesense style instant search-as-you-type with it, but it now also supports full vector hybrid search using this compressed BBQ stuff to make it super memory efficient. It’s really solid.
With Elasticsearch adding this, multi-vector embeddings are now officially mainstream.
looks like some multi vectors are cooking in #elasticsearch — you know, for col* github.com/elastic/elasti…
Metadrop @metadrop achieved a 10/10 score on their homepage using Dries Buytaert's HTTP Header Analyzer, writes @sanzante. This tool evaluates HTTP response headers, including Drupal-specific ones, providing recommendations for optimization. bit.ly/4eIOvqB
Now that the few post-merge failures are fixed, time to take a minute to celebrate the achievement: Lucene 10 will support decoupling search concurrency from the index's segment geometry. Still work to do, but a great step forward: github.com/apache/lucene/…
Elasticsearch (and Kibana) are Open Source, Again! Soooo excited. Read more in the blog I wrote elastic.co/blog/elasticse…
In part 4 of the ‘Inside Elastic’ blog series… Antonio Schönmann addresses the challenges, solutions, and outcomes of refining search effectiveness to perfect RAG search. Learn more in this blog → go.es.io/3Mlxt5D #RAG #Elasticsearch #GenerativeAI
Gotta brag on my teammates. Autoscaling is incredibly difficult to get right. However, we are making steps in the right direction. @elastic with zero systems management is pretty neato ;) elastic.co/search-labs/bl…
console1984: pretty cool feature in rails for protecting access to sensitive information in #elasticsearch
Our brand new podcast, “You Know, For Search” is live! Join host Steve Mayzak for technical discussions on #Elasticsearch and #Lucene, #RAG, semantic and keyword search, and so much more. New episodes dropping every other week! 🎧 Ep 1 with Serena Chou: go.es.io/4fDzdVt
Discover how scalar quantization can be used to reduce the memory footprint of vector embeddings in Elasticsearch: go.es.io/3Wm8bZN #ElasticSearchLabs #ML #Elasticsearch
if you need help, discuss.elastic.co or ela.st/slack are good places for questions :)
communityinviter.com
Join Elastic Stack Community on Slack - Community Inviter
Join Elastic Stack Community on Slack. Powered by Community Inviter. You will get an invitation soon. Check your inbox.
Our July #DevRel newsletter is out now! Read more for updates on… ✅ What’s new in Elastic’s distribution for @opentelemetry ✅ The latest content from the community ✅ Upcoming events & meetups Dive in now: go.es.io/3Wx9S88
I've been working on this for a while, so excited to see it out there gag.gl/Q7Ynwq
apache arrow's streaming format is coming to ES|QL (@elastic's piped query language). popular in data science and the de-facto standard for dataframe interchange with an efficient binary format that allows zero-cost deserialization github.com/elastic/elasti…
Dive into the different similarity techniques supported by Elasticsearch when it comes to searching for neighboring vectors and how to score them: go.es.io/3RYxOi2 #ElasticSearchLabs #Elasticsearch #VectorSearch
seeing all the drama around BM42 to find something cheaper than SPLADE: it doesn't have a fancy (or made up?) name but we've seen good improvements with token pruning in #elasticsearch: elastic.co/search-labs/bl…
Finally Deep Learning for Search is printed! Thanks to everyone that contributed! @ManningBooks @chrismattmann @deeplearning4j #lucene #dlearn
Explore the simplification of kNN search! k and num_candidates are now optional. But, picking good default values for them was a tricky undertaking. This blog explains how we got there: go.es.io/3RfDBPT #ElasticSearchLabs #kNN
We’re excited to introduce Playground! Iterate and build your RAG workflows by A/B testing LLMs, tuning prompts, and chunking data. Find out how this intuitive UI can help you quickly configure #semanticsearch for production: go.es.io/4cnSl7G
United States Xu hướng
- 1. Prince Andrew 23.3K posts
- 2. Ryan Clark 2,543 posts
- 3. $AMZN 47.2K posts
- 4. Sydney Sweeney 105K posts
- 5. Scream 7 41.1K posts
- 6. Rhule 5,222 posts
- 7. #Prop50VoteYes 4,324 posts
- 8. Happy Halloween 44.8K posts
- 9. Usha 31.5K posts
- 10. Somalia 58.6K posts
- 11. Necas 2,686 posts
- 12. $AAPL 82.2K posts
- 13. Andrew Mountbatten Windsor 5,381 posts
- 14. Poot 11.4K posts
- 15. Royal Lodge 5,709 posts
- 16. Bill Kristol 1,354 posts
- 17. Animal Crossing 30.9K posts
- 18. Buckingham Palace 5,777 posts
- 19. NextNRG Inc 2,198 posts
- 20. Heritage 41.8K posts
Something went wrong.
Something went wrong.