Complex tasks call for more than one perspective 🧠 Multi-agent systems go beyond a single LLM by combining specialized agents, each with their own tools, memory, and task. In this blog, Erika Shorten and Tony Kipkemboi unpack the core building blocks to CrewAI and

Your fitness data is living in silos. Your insights are nowhere. 𝗠𝗲𝘁𝗮𝗯𝘂𝗱𝗱𝘆 — built on 𝗪𝗲𝗮𝘃𝗶𝗮𝘁𝗲 — transforms this scattered health data into intelligent experiences that understand context, not just metrics. 🏋️♀️ Here's how they turned fragmented fitness data

We’re building the next chapter of NetBird's growth playbook with a focus on automation, data, and creativity. If you’re curious about how real startup growth works from the inside, we’re looking for a Growth Engineering Intern to join us in Berlin. You’ll help us shape and

🚀 What if landing your dream job felt less like an interview and more like a hackathon? That’s the idea behind Hack & Get Hired @ SF Tech Week We’re flipping the script: instead of résumés and endless calls, engineers will showcase real skills, build real automations, and walk

Think chunking is just "split text every 500 tokens"? That's why your RAG system can't find relevant information. It’s time to level up your chunking game 😎 Most developers jump straight to fancy retrieval techniques, but it’s really your chunking strategy that can make or

Turning this into the beginning of my learning journey into the whole "memory in agents" topic.

Static fitness dashboards are dead 💀 MetaBuddy is proving that the future of wellness tech is conversational AI powered by vector search across unified health data. Here's the problem right now: wellness data is fragmented across nutrition apps, workout trackers, sleep

This is how your agent's long-term memory is stored in Weaviate vector database

At The Sandbox we want to create the best environment for founders in their earliest stage. Part of that is to bring them together with people who did it before. Therefore we had Julian, Co-Founder of Berlin based Unicorn wefox as a guest for dinner. The full episode is now

Here's my hot take: if you're still building RAG 𝘄𝗶𝘁𝗵𝗼𝘂𝘁 𝗮 𝗾𝘂𝗲𝗿𝘆 𝗮𝗴𝗲𝗻𝘁, you're missing out on tons of performance. Query agents are the best thing you can do to easily improve your RAG performance. Why? Because users are messy. In the real world, user

I'm testing out different tools for the memory layer of AI agents. First up: mem0. Mem0 extracts memory information from conversations using an LLM and then allows you to CRUD over memories stored in your database, like Weaviate vector database. If you're curious, give it a spin in

If you're late to the whole "memory in AI agents" topic like me, I recommend investing 43 minutes to watch this video. In this video, Adam Łucek (sorry, couldn't find a handle) explains AND implements the 4 different types of memory from the CoALA paper: • working memory •

Agentic RAG used to require a team of engineers. Now? Visual workflows. We're building one live next week. The n8n.io Weaviate vector database node gives you 4 ways to work with data, from simple document insertion to full agentic workflows. Building RAG might be easy, but building it to

Bumble ist btw building with Weaviate vector database

8am on a Saturday morning in Berlin! Berghain or {Tech: Europe} Hackathon? Can not tell...

Biggest AI movement in Europe, just spitting some facts 🇪🇺

All nighter and truly believing in my team, making law cool again! Building European AI 🇪🇺 Bela Wiertz has no chance, my team is super cracked.

Your RAG system is slow because it's doing unnecessary work. REFRAG just proved that 99% of cross-passage attention is wasted computation. As context windows get longer, time-to-first-token latency explodes quadratically. 𝗥𝗘𝗙𝗥𝗔𝗚 is a newly released framework that

After Berlin, London, Paris, Munich & Stockholm - Copenhagen is the next city {Tech: Europe} is hosting a hackathon in. Join us, byFounders, OpenAI, Light & many others for a day of building: luma.com/copenhagen-hac… To unite Europe, you have to bring people from all ecosystems

Insurance claims that took days now take seconds. MarvelX built AI agents powered by Weaviate’s vector search which process travel-related claims in production. 𝗧𝗵𝗲 𝗽𝗿𝗼𝗯𝗹𝗲𝗺: Insurance companies are stuck with manual workflows that don't scale. Spreadsheets.