Raphael De Lio

Tbm pode ser esporádico

YouTube video by Yann LeCun Convolutional Network Demo from 1989 [restored version]

Nice! I came across LeCun's 1989 demo a couple of months ago: www.youtube.com/watch?v=H0oE...

YouTube video by Redis What is an embedding model?

Semantic caching is powered by embedding models which in turn are ML algorithms specialized in turning text into vectors that encode aspects of its semantics. By comparing these vectors we can estimate how semantically similar two pieces of text are. I explore it a bit further in the video:

The #Redis vector sets are really cool.

Just using basic movie information (e.g. "Title: <title>, Genre: <genre>, Year: <year>, Synopsis: <synopsis>" results in quite good similarity searches.

Also the ability to `SETATTR` and `GETATTR` and filter on them is awesome.

It is! Even got the OG controller!

That’s the first time I’m playing. Let’s see how far I can get. Did you lose the save?

Zelda the wind waker - One of the most beautiful video games I’ve played. Spent a couple of hours playing it today

I still haven’t touched my laptop today, but ended up watching a video from the Primeagen and my wife told me it classifies as programming 😅 I don’t agree though

Weather was better in the afternoon. I usually go for walks with my wife and dog. Just came back from a long one 😃

I was going to say: go outside. But then I checked the weather in NL. 😅

How many of us are out there? 👀

Saturday 10:30am, and I had the urge to open my MacBook and editor multiple times 😅

Told myself I won’t do any programming this weekend, but I don’t know if I can. I can’t really not think about it. I’m an addict, I need help.

✅ Exatamente! Geralmente um texto escrito por LLM vem com:

✏️ Uso razoavelmente correto da ortografia
📋 Listas
😘 Uso excessivo de emojis

🔥 Conclusão

Textos assim são forte indício do uso de LLM.

Quer um gráfico que mostra o uso de LLM’s dos anos 70 até os dias atuais?

Qualquer coisa só avisar!

Semantic Caching with Spring AI & Redis Build intelligent semantic caching with Spring AI and Redis to dramatically reduce LLM API costs. Cache responses by meaning, not exact…

Unlike traditional caching that requires exact query matches, semantic caching understands the meaning behind queries and can return cached responses for semantically similar questions

medium.com/redis-with-r...

cc @redis.io @danvega.dev @bsbodden.bsky.social @dashaun.com @starbuxman.joshlong.com

Gosto da zero

Join the Redis AI Challenge: $3,000 in Prizes! We are so delighted to partner with Redis for a new DEV challenge! Running through August 10, the...

Have you checked out the Redis AI Challenge yet? This super powerful multi-model platform is an open source in-memory data store and can be used as a:

🚩 cache
🚩 vector database
🚩 document database
🚩 streaming engine
🚩 message broker

Join today! 👇
dev.to/devteam/join...

hoje na minha última aula da Formação Plenos do #DevEficiente eu vou discutir sobre async processing com mensageria.

Talvez hoje os alunos(as) entendam parte dos meus traumas com mensageria, brokers e sistemas distribuídos 😅 Espero conseguir apresentar direitinho meus medos em detalhes, hehe.

Love it! I also came back to some physical magazines this year.

Por quê?

YouTube video by Redis Exact vs Approximate Nearest Neighbors in Vector Databases

Redis has JSON and search support since 2017. They were separate modules and with Redis 8 they’ve become part of its core.

Search indices is also under AI because vector search is built on top of searching algorithms such as HSNW and FLAT.

I happen to have a video about it: youtu.be/9NvO-VdjY80

Technology | 2025 Stack Overflow Developer Survey

survey.stackoverflow.co/2025/technol...

📷 Redis is leading in AI. It’s officially the most popular database for AI agent data storage.

Redis has been a vector database since 2022 and with Redis 8 we’ve taken a big leap improving the query engine for better performance and adding the new Vector Sets data structure! #vectordatabase

🔴 Redis remains one of the most used databases, growing 8% compared to last year. Even better, most people who use Redis admire it — and about a quarter of those who don’t yet use it want to start.

And besides that...

Stack Overflow’s 2025 Developer Survey just dropped, and it’s packed with insights on the state of software development.

Two things stood out to me (especially since I work at Redis) 🧵 cc @redis.io

LLM Embeddings Explained: A Visual and Intuitive Guide - a Hugging Face Space by hesamation This app explains how language models transform text into meaningful representations through embeddings. It provides a visual guide to help you understand traditional and modern language model tech...

huggingface.co/spaces/hesam...

It still surprises me how often people talk about LLMs without even mentioning word embeddings. Without embeddings, LLMs wouldn’t even exist.

Hesam Hassani does a fantastic job breaking down what embeddings are (not just technically, but also from a historical perspective) in this article.

Também nunca precisei fazer

Before and after joining @redis.io

The takeaway? Temperature isn’t a “creativity dial.”

LLMs can produce outputs that look creative (poems, jokes, plot twists) but they’re remixing patterns, not inventing.

👉 When AI surprises us, does that count as creativity?Or are we just giving randomness too much credit?