Hey there,

What a fantastic year it has been! In 2023, the world of AI saw tremendous progress, making it an exciting time for learning and growth. As we wind down and approach the new year, it's a good moment to take a breather, look back on the past year, and appreciate the highlights before gearing up for what's ahead. We've taken some time to review the noteworthy moments that made 2023 special for Haystack and its community.

A big thank you for being part of this journey! Here's to a wonderful New Year filled with dizzying generative AI and NLP advancements 🥂✨

💫 Haystack Highlights

• The biggest highlight of the year was Haystack 2.0-Beta, which is a more flexible, extendable, customizable version of Haystack v1 💯 
• Haystack surpassed 12,000 stars on GitHub 🌟
• It has been exactly 2 years since the first stable release of Haystack 🎉
• We launched Haystack Integrations to showcase external packages by the community and additional technologies that can be used with Haystack 🧩
• We hosted the first Advent of Haystack 🎄
• Over 80 pull requests from the community were merged 🚀
• We launched Haystack Blog, where we publish articles about Haystack, LLMs, demos, and use cases 📚 P.S. Did you know that you can contribute to Haystack Blog?

👥 Community Highlights

• Our community has reached 2300+ members on Discord, join the Haystack Discord community to ask questions, stay informed about Haystack updates, and connect with fellow Haystackers 🤖
• Discord has been the platform for numerous office hours, a handful of live coding sessions, and engaging Road to 2.0 sessions  🛣️
• We proudly participated in Hacktoberfest for the first time and welcomed new Haystack contributors 🎃
• We introduced our monthly Haystack community newsletter that you're reading right now 🗞️
• 120+ Advent of Haystack participants took the time to give us feedback on Haystack 2.0-Beta 🎄
• We participated in the Anthropic Hackathon 🧑‍💻
• Haystack has spread its wings, follow us on Twitter/X at Haystack_AI 🐦

📆 Webinar Highlights

We collected some of the webinars that the Haystack community found interesting in 2023:

• Prompting LLMs Using Haystack

• Building Applications with LLM-Based Agents

• From Hybrid Retrieval to RAG with OpenSearch and Haystack

• Optimizing LLMs with Retrieval Augmented Generation and Haystack 2.0

📖 Blog Post Highlights

Here are the most visited blog posts of 2023:

• Gemini Models with Google Vertex AI Integration for Haystack: Multi-Modal components and function calling with the new Gemini integrations for Haystack 2.0

• The Beginner's Guide to LLM Prompting: Engineer your natural language prompts to get the best answers from a large language model

• Deploying RAG to Production: From a working prototype to a RAG AI system in production: an overview for data scientists

• How to Prevent Prompt Injections: Prevent prompt injections leveraging our new open-source model and dataset

• Enhancing RAG Pipelines in Haystack: How the latest rankers optimize LLM context window utilization in RAG pipelines

👏 Some of Your Work

In 2023, your contributions to the community were truly outstanding, and we want to highlight some of these works that we enjoyed seeing:

• @cclauss, @faaany, @danielbichuetti, @awinml, @zoltan-fedor, @benheckmann, @nickprock and @recrudesce contributed the most with their PRs
• @AIAnytime on YouTube delivered two invaluable videos on Haystack: 
  

  Chat with Data App: RAG using Mistral 7B, Haystack, and Chainlit and RAG Implementation using Mistral 7B, Haystack, Weaviate, and FastAPI

• Laura Funderburk hosted a mentorship program to help users build with Haystack and open source projects, in addition to her great Haystack demos

and numerous members showcased their projects with the community on the #show-and-tell channel of Discord, as well as on Linkedin and Twitter 🤩
 

🔜 Next

• We will release the stable version of Haystack 2.0 in Q1 2024. Check out Haystack public roadmap to see the upcoming features. 
• We have just created a new repository: Haystack Cookbook. We will extend this collection with more example notebooks in 2024.

🥇 New with Haystack 1.23

While 2.0-Beta is out, we continue to support the stable release of Haystack 1.x 💪 Check out the full release notes to see the whole scope of the new Haystack release. But here are some highlights:

• ⛰️ Support for Amazon Bedrock models! Simply provide the name of the models, like Llama-2-70b-chat, in the PromptNode to start using them.
• 🗺️ MongoDB Atlas integration with MongoDBAtlasDocumentStore.
   

💙 deepset Highlights

deepset, the company behind Haystack, announced a $30 million funding round led by Balderton Capital, with participation from existing investors GV, Harpoon, System.One, and Lunar. Read more about it here.

🤝 Let us know what you think

We would love to hear what the community thinks about Haystack. Please leave us a comment on G2 💙

Lots of love,

Developer Relations team at deepset 💙