Nancy Ntinyari’s Post

𝗖𝗵𝗲𝗰𝗸 𝗮 𝗗𝗶𝘀𝗿𝘂𝗽𝘁𝗶𝘃𝗲 𝗢𝗽𝗲𝗻 𝗦𝗼𝘂𝗿𝗰𝗲 𝗟𝗟𝗠! 💡 In the past few days, the entire tech world has been excited about a new large language model released by DeepSeek, a chinese AI company founded in 2023. DeepSeek-R1 has a number of notable features listed below: ✅ Open Source: in contrast to other leading models, DeepSeek-R1 is publicly available on Github and HuggingFace. ✅ State Of the Art: according to the company, DeepSeek-R1 has comparable performance to OpenAI-o1 on various benchmarks. ✅ Efficient: along with the base model, DeepSeek has released smaller ones that also perform exceptionally well. ✅ Diverse: the DeepSeek-R1 model is extremely capable on math, coding, reasoning and various other tasks. I have tested the new model with challenging questions, including unsolved math problems like the Riemann hypothesis, and I am quite impressed with the results! Do you think that DeepSeek will disrupt the global AI industry? Check the links below for more information, and make sure to follow me for regular data science content. 𝗗𝗲𝗲𝗽𝗦𝗲𝗲𝗸-𝗥𝟭 (𝗚𝗶𝘁𝗵𝘂𝗯): https://lnkd.in/dE_svveG 𝗟𝗲𝗮𝗿𝗻 𝗠𝗟 𝗮𝗻𝗱 𝗧𝗦 𝗙𝗼𝗿𝗲𝗰𝗮𝘀𝘁𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗣𝘆𝗖𝗮𝗿𝗲𝘁📚: https://lnkd.in/dyByK4F #datascience #python #machinelearning #deeplearning

Version 4 of Hopsworks' #ai #lakehouse is a milestone release. A lot of work went into this, and I think it will be truly transformative in how organisations can build AI systems (not just "yet another model") in the cloud or on-prem.

Today we have exciting news from Hopsworks! Over the summer, we launched Hopsworks 4.0, and it’s now generally available! This release brings the AI Lakehouse to the forefront by extending traditional lakehouses support for real-time AI and LLM systems. Plus, we’re making Python a first-class citizen with our leading AI Query Engine, powering the Feature Store. Our mission remains the same; help teams, big and small, build and operate smarter AI-powered applications—both batch and real-time—more easily than ever. And there is more to come! As we roll out the 4.0 release globally, new innovative features and technologies will be introduced in the coming months, making workflows faster and simpler to create. Learn more about how we’re extending the Lakehouse to meet the demands of AI in our latest blog: https://lnkd.in/dji3ZdmM I’ll also be hosting a short webinar in a few hours where I’ll walk through the key aspects of this new paradigm. Register here: https://lnkd.in/dB_TDBS4

This weekend, I built a Research Assistant ChatBot using LangChain, Streamlit, and GPT-4. The bot leverages Retrieval Augmented Generation (RAG) to analyze PDFs, Excel, and CSV files, delivering precise, real-time answers from your research papers and data. What It Offers: - Multi-Document Support: Handles various file formats seamlessly. - AI-Powered Insights: GPT-4 provides context-aware answers. - User-Friendly Interface: Features a Streamlit interface with a dark mode option. - Real-Time Performance: Offers instant responses as data is processed. This project aims to simplify research processes by turning complex documents into actionable insights within seconds. If you work with large datasets or research papers, this tool can enhance efficiency. Explore the code on my GitHub here: [https://lnkd.in/gMNhCbN8]. #RAG #GPT4 #LangChain #Streamlit #DataScience #AI #Research #SeniorDataArchitect #Python #GenerativeAI #Data

Global Data & AI Virtual Tech Conference We are thrilled to welcome Aleksander Molak as a speaker at the upcoming Global Data & AI Virtual Tech Conference on January 18, 2025! As a Causal ML Tutor at the University of Oxford, Aleksander is a leading expert in causal inference and machine learning, with extensive experience working with Fortune 100 Coaches and Fortune 500 Companies across Europe, the USA, and Israel. His session titled “Causal Inference and Discovery: Unlocking The Secrets of Modern Causal Machine Learning With #Python,” Aleksander will delve into advanced techniques for making better decisions through causal inference. He will explore how to navigate challenges in causal modeling, focusing on methods that allow for valid inferences even under violated assumptions. Aleksander is the author of the best selling book Causal Inference and Discovery in Python, which explores the intersection of causal inference and AI. His mission is to democratize causality for businesses and practitioners, making complex concepts accessible and actionable. He also founded Causal Python.io , providing training and consulting in causal inference for corporate teams.PCausa Join us to hear Aleksander share his insights on modern causal machine learning techniques that can unlock powerful decision making capabilities. This is an incredible opportunity to learn from one of the foremost experts in the field! RSVP for free here: https://lnkd.in/d2avHMJs #AleksanderMolak #CausalInference #data #machinelearning #artificialintelligence #bigdata #analytics #technology #future #innovation #career #productivity #education #AI #instagood #techevent #techconference #dataglobalhub #GDAI #creativity #pic #viral #trending

As data professionals, we pour hours into learning new tools (Python libraries, anyone?), refining models, and digging for insights. But sometimes, the payoff seems distant. This image depicts the 'Plateau of Latent Potential,' a concept from "Atomic Habits" that resonates deeply with the data journey. We often face a 'Valley of Disappointment' where our efforts don't translate into visible results. But the key is consistent effort – building those foundational skills and knowledge. Eventually, we reach the 'Plateau,' where our accumulated work fuels exponential growth. Think of mastering feature engineering: tedious at first, but transformative for model performance later. Keep pushing, data friends! The breakthrough is coming. #DataScience, #Analytics, #MachineLearning, #AI, #DataVisualization #Motivation

🎉 Unlock the Power of AI with RAGFlowChain! 🎉 I'm beyond excited to introduce RAGFlowChain, the next-generation Python toolkit that’s set to transform the way you build and deploy AI models. Whether you're a seasoned data scientist or a cutting-edge AI developer, this is your go-to library for creating powerful Retrieval-Augmented Generation (RAG) pipelines. 💡 Why RAGFlowChain? - 📚 Comprehensive Data Handling: Seamlessly load and process data from books, news articles, YouTube videos, websites, and local documents—all into a single DataFrame. - 🔍 State-of-the-Art Vector Databases: Effortlessly create and manage vector databases with Chroma, FAISS, and other top-tier methods. - 🤖 Advanced RAG Chain Management: Easily integrate retrieval with generative models to build intelligent, context-aware AI systems. - 🛠️ Customizable & Extensible: Tailor the pipeline to meet your exact needs, whether you're working on research, product development, or AI-powered insights. RAGFlowChain is designed to make your AI workflows smarter, faster, and more efficient. Perfect for anyone looking to push the boundaries of AI-driven innovation. 🚀 Ready to get started? - PyPI: [Explore on PyPI](https://lnkd.in/ebd5pvud) - GitHub: [Check out the code](https://lnkd.in/exhQAN6z) 💬 Your feedback and contributions are crucial! Let's collaborate to refine and enhance this tool, making it even more powerful for the entire AI community. Together, we can revolutionize how AI interacts with data. Let’s build the future, one RAG pipeline at a time! 🌟 #RAGFlowChain #AI #MachineLearning #Python #DataScience #Innovation #OpenSource #LangChain #Chroma #AItools #GenerativeAI #RAG

🚀 Excited to share that I’ve completed a course on Customizing Open-Source AI Models using LLaMa (Large Language Model Meta AI)! This hands-on learning experience covered LLaMa's architecture, prompting, deployment, and training using Python notebooks. It's incredible to see how adaptable LLaMa is for real-world use cases in both enterprise and startup environments. Looking forward to applying these skills in AI/ML projects and driving innovation! 🌟 #AI #ML #LLMs #LLaMa #OpenSource #Learning

Unlock the Future of AI with LangChain! 🚀 Struggling to keep up with the rapid advancements in AI? Meet your new guide, "LangChain Python for RAG Beginners"! This comprehensive book takes you from the basics to building sophisticated AI agents using LangChain. Whether you're a seasoned programmer or a newcomer, this book has something for everyone. Why You Should Read This Book: 1. **Transformative Learning**: Understand and harness the power of AI agents that can automate tasks, make decisions, and provide meaningful insights. 2. **Hands-On Projects**: Follow along with 50+ code examples to build practical applications, culminating in a fully functional Conversational AI Agent. 3. **Future-Proof Your Skills**: Stay ahead in the competitive tech landscape by mastering the integration of AI agents into real-world applications. What’s Inside? - **Chapter 1**: Introduction to LangChain - **Chapter 6**: Automating Tasks with Chains - **Chapter 7**: Adding Memory to Your Applications - **Chapter 13**: Building Agent Web User Interfaces using Streamlit …and much more! With over 30 years of experience in the IT industry, Karel Hernandez Rodriguez guides you through this transformative journey, ensuring you walk away with the knowledge and tools to create AI agents that can revolutionize workflows and drive innovation. Ready to dive in? Get your copy now and transform your AI projects! 👉 [Buy Now on Amazon](https://lnkd.in/gCTV6hB2) #AI #LangChain #MachineLearning #Python #BookRelease #TechInnovation #AIDevelopment #ConversationalAI #Streamlit #GenerativeAI #TechSkills #FutureOfWork Don’t miss out on this essential guide to the future of AI development!

McKinsey releases a new report that dissects how the explosive growth generative is leading to a significant capacity shortfall due to the growing demand of AI data centers, which at this rate would require tripling in capacity by 2030: An interesting analysis diving into how this surge has been driven primarily by hyperscalers hosting advanced AI workloads, and which limit computational resources for production machine learning practitioners - it is clear that the industry will need to adapt strategies for efficient resource utilization. Report: https://lnkd.in/eHHcM_U5 -- If you liked this article you can join 60,000+ practitioners for weekly tutorials, resources, OSS frameworks, and MLOps events across the machine learning ecosystem: https://lnkd.in/eRBQzVcA #ML #MachineLearning #ArtificialIntelligence #AI #MLOps #AIOps #DataOps #augmentedintelligence #deeplearning #privacy #kubernetes #datascience #python #bigdata