Exploring Agentic RAG with Llama 3.2 (Link below 👇) Stefano Fiorucci from deepset recently published a cook-book on using Agentic Retrieval Augmented Generation (RAG) with the new Llama 3.2 model. 🎯 Here’s the approach: Answer questions from a knowledge base Use Web search as a fallback if more context needed Tech stack: 🏗️ Haystack for orchestration 🦙 Llama-3.2-3B-Instruct 🌐 DuckDuckGo API Check out the full example in the Haystack Cookbook: https://lnkd.in/gUbxvQ6b #LLama3.2 #NLP #AI #LLM #RAG #AgenticRAG
deepset
Softwareentwicklung
Accelerate your AI strategy – not your costs. We help you apply LLMs to your business the way you need them.
Info
For teams in need of optimizing their time and resources, our enterprise LLM platform, deepset Cloud, lets you design, securely deploy, and monitor your AI applications leveraging pre-built components and a unified development environment. Since 2018, we have enabled organizations across the financial, manufacturing and legal industries to unlock the potential of natural language processing for their business use cases. If you can dream it, you can build it with deepset Cloud. Since 2018, deepset has helped AI teams to apply the latest in NLP to real-world applications. We also provide Haystack — a trusted, reliable open-source technology to build standalone applications with LLMs.
- Website
-
https://www.deepset.ai/
Externer Link zu deepset
- Branche
- Softwareentwicklung
- Größe
- 51–200 Beschäftigte
- Hauptsitz
- Berlin
- Art
- Privatunternehmen
- Gegründet
- 2018
- Spezialgebiete
- Artificial Intelligence, Natural Language Processing, Question Answering, Large Language Models, AI, LLMs, Machine Learning, Semantic Search, Conversational AI, Question Answering, Generative AI, Information Extraction, vetor-based search und Document Similarity
Orte
-
Primär
Wallstraße 9-13
c/o Spaces Spindlershof
Berlin, 10179, DE
-
80 Broad St
5th Floor
New York, NY 10004, US
Beschäftigte von deepset
-
David S. Batista, PhD
NLP Engineer | Machine Learning Engineer | Software Developer
-
Jay Wilder
Marketing Leader with a Deep Love for the Startup to Enterprise Journey
-
Vladimir Blagojevic
Senior Software Engineer @ deepset
-
Iván López
♾ DevOps Engineer | ⚗ Kubernetes Alchemist | ⛅ Cloud Seeder | Join us in building amazing NLP solutions @ deepset 🚀
Updates
-
Do LLMs benefit from longer context windows as opposed to RAG with short chunks of text? We have created a demo so you can see the difference for yourself! Compare Basic RAG and Long Context Language Models (LCLMs) side by side. Read full post ⏩https://lnkd.in/gqWPi2td Try the demo ⏩ https://lnkd.in/ghd9TnET
-
You can talk for hours about how great your AI framework is. But the ultimate proof is when real users put your framework through its paces. That's what Josh Longenecker did by building several real-world LLM applications with Haystack. He's concluded that Haystack is the gold standard for LLM orchestration out there, thanks to these five benefits: ⭐ Comprehensive documentation ⭐ Modularity ⭐ Customizability ⭐ Serialization capabilities ⭐ Production readiness Want to know more? Check out Josh's in-depth article on Medium! https://lnkd.in/g2NdgxGS
-
Only Custom RAG Is Production-Ready RAG Read the full article here: https://lnkd.in/eJMvxpTc If you’re interested in generative AI, you’ve probably seen a retrieval augmented generation (RAG) pipeline before. But did you know that there are at least four points in a basic RAG setup that you can (and probably should!) customize: Before retrieval, during retrieval, after retrieval, and again after LLM generation. Compound AI sets the stage for generative AI systems that can be tailored to unique use cases. deepset Cloud lets you iterate and customize with quickly and easily. Stay tuned for examples of the most common custom RAG setups 😎
-
🚀 Missed our webinar? [link to recording below 👇] In this session, we dive into how Language Learning Models (LLMs) can transform the way you interact with your data—allowing you to ask your Snowflake database questions in plain language, just like talking to a colleague. You'll learn: - Why LLMs struggle with tabular data (and how to fix it) - How LLMs can bridge data gaps - Tips to improve how models understand your questions - Real-world examples of companies using AI to get clearer insights Don’t miss out—catch the recording now and discover how to build an AI layer on your Snowflake data! ⬇️ https://hubs.li/Q02L73Lv0 #Snowflake #LLM #AI #DataScience #BusinessIntelligence #Webinar
-
💡 Enhance your AI Systems with Custom RAG 💡 To truly harness the power of generative AI and gain a competitive advantage, you must customize and extend the basic setup to meet your specific business needs. The difference between a generic solution and a customized, production-ready RAG system is the key to unlocking the value of your AI initiatives. Check out our latest blog post on how to extend and customize basic RAG systems for maximum impact: https://hubs.li/Q02QLYjb0
Customizing Retrieval Augmented Generation (RAG) Systems
deepset.ai
-
Happening Tomorrow 👉 https://lu.ma/dky2hvoj Tuana Çelik, Developer Relations Lead at deepset, is hosting a livestream with John Gilhuly from Arize AI, diving into tracing for AI applications. If you're building AI pipelines or want to learn how to trace your Haystack workflows using Arize Phoenix, this session is for you . Here's what to expect: - Understanding tracing in the AI development cycle - Setting up tracing for Haystack using Arize Phoenix - Open Q&A 🕕 When? 12:00 PM ET / 6:00 PM CET #AI #DataScience #Haystack #ArizePhoenix #MachineLearning #AITracing
-
🚀 Successful AI projects need the right team! Whitepaper below 👇 Is your organization struggling to implement AI solutions? The challenge might not be the technology itself, but the collaboration between your data science, development, and executive teams. We’ve got you covered with our latest whitepaper: "Leading a Successful AI Team" 📄👇 🎯 Key Takeaways: - Essential Skills: Identify the key expertise each member of your AI team needs. - Leadership Insights: Discover the leadership styles that drive innovation. - Collaboration Strategies: Learn how to unite data science, development, and the C-suite for seamless, product-driven AI projects. https://hubs.li/Q02M8Nrm0 #AI #Leadership #DataScience #SoftwareDevelopment #AILeadership #Whitepaper #TeamCollaboration #Innovation #ArtificialIntelligence
How your organization can realize the full potential of AI with the right team in place
landing.deepset.ai
-
🌐 Our Co-Founder and CTO Malte Pietsch will be delivering a talk on Composable, Transparent, and Robust AI Applications at John Snow Labs's 2024 Online NLP Summit on September 24 at 3:10pm ET. ⚡ Don't miss this chance to learn about all things Generative AI, NLP, and Open Source, and ask Malte questions live! Register here for free: https://lnkd.in/dnS8eFjb
-
Download link below 👇 Did you know that 45% of organizations have seen an increase in AI investment since ChatGPT was released, and 68% of executives believe that generative AI benefits outweigh its risks. However, despite this excitement, organizations struggle to get started: Only 19% of organizations are currently running a pilot or producing generative AI. Read more in this report from Gartner®. Get the report here: https://hubs.li/Q02j-N4j0