Femke Plantinga

Traditional (SQL) databases rely primarily on keyword-based searches to retrieve information. These searches match the exact words or phrases in your query to the text stored in the database. While effective for many applications, this method has limitations when it comes to…

Built a RAG system? 🎉 Congrats - you’re already a context engineer (even if you don’t call yourself one yet) What we call "Retrieval-Augmented Generation" is never just about retrieval. It’s about giving our LLMs access to new relevant information, and more importantly, the…

Get started with Gemini Embedding using @weaviate_io, which supports over 100+ languages and flexible dimensions for performance and storage needs. Check out the notebook. github.com/weaviate/recip…

we got poached by Meta, but we consider any counteroffer @bobvanluijt

How do you feed a 500,000-word novel to an AI? You can't. Not all at once, anyway. This is where 𝗰𝗵𝘂𝗻𝗸𝗶𝗻𝗴 comes in. 𝗪𝗵𝗮𝘁 𝗶𝘀 𝗖𝗵𝘂𝗻𝗸𝗶𝗻𝗴? Chunking is the pre-processing step of splitting texts into smaller pieces of texts, i.e. "chunks". Each chunk becomes the…

Skincare recs are broken. We fixed them. If you’ve ever tried to build a routine and ended up with irritation, confusion, or ten tabs of conflicting advice… well, then you know the struggle. Too many apps give you generic recs based on bestsellers, vague “skin types,” or…

𝘎𝘦𝘮𝘪𝘯𝘪 𝘌𝘮𝘣𝘦𝘥𝘥𝘪𝘯𝘨𝘴 🤜🤛 𝘞𝘦𝘢𝘷𝘪𝘢𝘵𝘦 We just published a new recipe notebook showing you how to use Google's Gemini embeddings with Weaviate. • 𝟭𝟬𝟬+ 𝗹𝗮𝗻𝗴𝘂𝗮𝗴𝗲𝘀 𝘀𝘂𝗽𝗽𝗼𝗿𝘁𝗲𝗱: Build truly multilingual search experiences • 𝗠𝗧𝗘𝗕…

Multi-vector embeddings like ColBERT and ColPali are amazing for retrieval quality. But they come with a massive headache: memory costs that can explode your infrastructure budget. Here's how MUVERA solves this problem by converting multi-vector embeddings into single…

💡 New in Weaviate Console: Use the Query Agent directly! 👨‍💻 In this quick demo, I upload and automatically vectorize a bunch of websites. 🤖 The Query Agent then generates three smart queries — combining semantic search and filters — and runs them automatically. 👉 Try it…

Your AI chatbot is not dumb. It's all about prompt engineering - and yes, it's more of an art than a science. 𝗪𝗵𝗮𝘁 𝗶𝘀 𝗽𝗿𝗼𝗺𝗽𝘁 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴? Prompt engineering is the art and science of crafting the perfect input to get the best output from language…

Your AI chatbot is not dumb. It's all about prompt engineering - and yes, it's more of an art than a science. 𝗪𝗵𝗮𝘁 𝗶𝘀 𝗽𝗿𝗼𝗺𝗽𝘁 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴? Prompt engineering is the art and science of crafting the perfect input to get the best output from language…

Your AI chatbot is not dumb. It's all about prompt engineering - and yes, it's more of an art than a science. 𝗪𝗵𝗮𝘁 𝗶𝘀 𝗽𝗿𝗼𝗺𝗽𝘁 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴? Prompt engineering is the art and science of crafting the perfect input to get the best output from language…

RAG Benchmarks! Evals, evals, evals! ⚖️🚀 I am SUPER EXCITED to publish the 124th episode of the Weaviate Podcast featuring Nandan Thakur (@beirmug)! 🍻🔎 Evals continue to be one of the hottest topics in AI. Few people have had as much of an impact on evaluating search as…

The future of AI isn’t just text - it’s multimodal. If you know these 6 terms, you’ll be ahead of most developers building text-only systems. 𝗠𝘂𝗹𝘁𝗶𝗺𝗼𝗱𝗮𝗹 𝗥𝗔𝗚: weaviate.io/learn/knowledg… 𝗠𝘂𝗹𝘁𝗶𝗺𝗼𝗱𝗮𝗹 𝗲𝗺𝗯𝗲𝗱𝗱𝗶𝗻𝗴 𝗺𝗼𝗱𝗲𝗹𝘀: weaviate.io/learn/knowledg……

Most developers optimize their vector databases. Smart developers optimize their queries first. Here are 4 powerful techniques to optimize your queries before they hit your vector database: 𝟭. 𝗤𝘂𝗲𝗿𝘆 𝗗𝗲𝗰𝗼𝗺𝗽𝗼𝘀𝗶𝘁𝗶𝗼𝗻 Break complex questions into smaller,…

Think smart retrieval ends at semantic similarity? Well, Graph RAG adds the layer that connects the dots in your data! When you ask a question to a RAG system, you're not just relying on the LLM; you're depending on how that system retrieves context in the first place. In most…

“You don’t need RAG, context windows are huge now…” Sounds great, right? But the reality is: 𝗯𝗶𝗴𝗴𝗲𝗿 𝗶𝘀𝗻’𝘁 𝗮𝗹𝘄𝗮𝘆𝘀 𝗯𝗲𝘁𝘁𝗲𝗿, especially if it’s slower, costlier, and oblivious to updates. Join @Box’s CTO Ben Kus, along with @bobvanluijt and @CShorten30, in…

Vector indexes are the difference between waiting 10 seconds and 10 milliseconds. Here’s what most developers don’t know about them: 𝗪𝗵𝗮𝘁 𝗶𝘀 𝘃𝗲𝗰𝘁𝗼𝗿 𝗶𝗻𝗱𝗲𝘅𝗶𝗻𝗴? In vector databases, these (vector) indexes organize your embeddings to make similarity searches…

Is your RAG system connecting the dots, or just finding them? Here’s why GraphRAG might be the missing piece in your AI retrieval system: 𝗧𝗿𝗮𝗱𝗶𝘁𝗶𝗼𝗻𝗮𝗹 ("𝗡𝗮𝗶𝘃𝗲") 𝗥𝗔𝗚 • Uses semantic search to find relevant documents based on meaning • Each piece of content…

New integration: Cleanlab + @weaviate_io! Build and scale Agents & RAG with Weaviate. Then add Cleanlab to: - Score trust for every LLM response - Flag hallucinations in real time - Deploy safely with any LLM 📷weaviate.io/developers/int…

Not all embedding models are created equal. Here’s HOW AI understands different types of data: There are two main approaches to embedding models. 👇 𝗨𝗻𝗶𝗺𝗼𝗱𝗮𝗹 𝗘𝗺𝗯𝗲𝗱𝗱𝗶𝗻𝗴𝘀: Unimodal embeddings represent a SINGLE modality (text, image, audio, etc.) in a…