Elevate your AI applications: Launching Verba 0.3.0, your customizable open source RAG framework

Hello friends, 🤗

Welcome back! November has been transformative, marked by the release of Verba 0.3.0—a customizable RAG framework. Learn more in the newsletter! Plus, prepare for our global end-of-year tour covering Berlin, Amsterdam, Paris, San Francisco, and London. Let's dive in!

Verba 0.3.0

We're excited to introduce Verba - our open-source RAG framework. Designed to keep your data private and completely free, Verba uses open-source models such as SentenceTransformers. The best part? You can run Weaviate locally. Find out more in our GitHub repository.

Uploading data with Verba's user-friendly front end has never been easier. We've integrated with unstructured data management tool UnstructuredIO to ingest your PDFs and GitHub to read and import files directly from your repos. You can also read files right from your filesystem.

****To ensure the highest accuracy in data retrieval, Verba offers chunking capabilities. You can decide how to chunk your data, at the word or sentence level, and also decide on the overlap of chunks for optimal data segmentation.

Choose embedding options from providers like OpenAI (AdaEmbedder), Cohere (CohereEmbedder), and Hugging Face (MiniLMEmbedder).

Select language models from Cohere, OpenAI, and Hugging Face to generate answers from data chunks.

Learn more

• Explore the Verba release by Edward Schmuhl on the official website and watch the release video for additional details

• For a detailed technical overview and tutorial for running Verba locally, check out Mert Cakmak's post on Medium

Vector data tips, tricks, and tech

How-to

• ✍️ An Overview on RAG Evaluation by Erika Cardenas and Connor Shorten

  • Learn about new trends in RAG evaluation and the current state of the art

• ✍️ Building an AI-Powered Shopping Copilot with Weaviate by Alea Abed and Zain Hasan

  • UK-based startup Moonsift is harnessing the power of AI with Weaviate

Explore AI and Vector Databases

• Discover our range of online workshops on AI, vector databases, and more at Weaviate's Online Workshops page

• Consider enhancing your understanding with the "Vector Databases: from Embeddings to Applications" short course from DeepLearningAI

Newest AI & data technology developments

Podcasts

• 🎙 RAGAS with Jithin James, Shauhl Es, and Erika Cardenas

  • We explore RAGAS, a rising star in Retrieval-Augmented Generation, discussing the innovative RAGAS score, optimization tips, and future directions such as fine-tuning smaller LLMs and exploring long-context RAG

• 🎙 Patrick Lewis on Retrieval-Augmented Generation - Weaviate Podcast #76

  • Patrick Lewis, lead author of the "Retrieval-Augmented Generation" paper, discusses the origin and architectural nuances of RAG, its impact on few-shot and zero-shot task performance, along with future directions exploring privacy, task-aware retrieval, and innovative text editing applications

Upcoming events

Weaviate World Tour: End of Year Special

Welcome to our Weaviate World Tour: Year-End Special! We love community and want to celebrate the end of the year together with you.

Let's explore the practical side of AI applications and vector databases together.

Whether you're an experienced developer or just starting, join us to learn how to build real-world AI solutions. This Weaviate End of the Year Tour is beginning a new era of community events.

Scheduled events:

• Dec 6 - Berlin: Event Details

• Dec 8 - Paris: Event Details

• Dec 12 - San Francisco: Event Details

• Dec 12 - London: Event Details

• Dec 14 - Amsterdam: Event Details

Stay connected with us on social media for updates, behind-the-scenes glimpses, and maybe even a surprise or two. Remember to check out the provided links for all the details on how to sign up. We can't wait to meet you!

On the calendar

• AWS re:Invent 2023 | Las Vegas, US | Nov 27 - 30, 2023

  • Visit us at booth #1620 at AWS re:Invent to learn how to simplify building AI-native applications with our open-source vector database

• AI Hack Party | San Fransisco, US | Dec 2, 2023

  • Join us in Hayes Valley, San Francisco, for an unparalleled hackathon experience where the future of technology meets the spirit of a party

Team updates

Join us and be part of the future of search

At Weaviate, we are committed to our values - the foundation of our company. They guide our decisions in better ways for our people, business, and future. If you share our passion for innovation and excellence, join our vibrant team as we're currently hiring for the role of Client-Focused Site Reliability Engineer (US-based).

Apply now! Let's build the future together! 🚀

Welcoming new faces

• Albert Adeseye joins us as a Technical Success Manager

• Aarthi Iyer joins us as a QA Engineer

• André Mourão joins us as a Software Engineer - Search

• Mohamed Shahin is joining us as our new Technical Success Manager

Thanks for reading!

Questions about Weaviate, vector databases, or anything else? Dive into the details on the Weaviate Forum. For a more direct and focused interaction, connect with the Weaviate community on Weaviate Community Slack.

Till the next one,

Femke