👋 Hello ,

📢 Welcome toDataPro #128~ Your Weekly Dose of Data Science & ML Innovation!

The world of AI, machine learning, and data science never slows down, and neither do we! This week’s edition is packed with breakthroughs, must-know tools, and career insights to keep you ahead of the curve.

🔹 Data & ML Reads: Explore Python Machine Learning By Example, Power BI mastery, deep reinforcement learning, and high-performance data manipulation with Polars.

🔍 Fresh Insights: A 27-day AI coding experiment, deep dive into LLMs, and why data scientists should embrace Docker.

🚀 Tech Trends: Advanced Time Intelligence in DAX, Multimodal search with BLIP-2 & Gemini, and Sparse Autoencoders in LLMs.

🛠️ ML Tool Showdown: Discover MoBA’s new attention mechanism, Microsoft’s Magma AI for robotics & UI, and Mistral Saba’s breakthrough in Arabic & Tamil NLP.

📊 Success Stories: Free interactive data visualizations with Marimo, SQLite-powered RAG, and how Decision Intelligence is shaping the future of data.

💡 Your AI & ML Knowledge Hub is Here! Dive into these game-changing trends, tools, and innovations.

🔗 Read it all now! ⬇️

Cheers,

Merlyn Shelley

Growth Lead, Packt

❯❯❯❯ Python Machine Learning By Example: Written by Yuxi (Hayden) Liu, Python Machine Learning by Example, Fourth Edition is a hands-on guide covering NLP transformers, PyTorch, computer vision, and deep learning. It emphasizes best practices for building and improving real-world machine learning models using Python.

❯❯❯❯ Microsoft Power BI Cookbook: Written by Greg Deckler and Brett Powell, Microsoft Power BI Cookbook (3rd Edition) is a detailed guide for data professionals, covering data integration, Hybrid tables, scorecards, real-time processing, governance, security, and advanced visualization. With step-by-step techniques, it helps you transform raw data into actionable insights using Power BI’s latest innovations.

❯❯❯❯Modern Time Series Forecasting with Python: Written by Manu Joseph and Jeffrey Tackes, Modern Time Series Forecasting with Python (2nd Edition) is a detailed guide for data professionals, covering machine learning, deep learning, transformers, probabilistic forecasting, feature engineering, and ensemble methods. With hands-on techniques, it helps you build, evaluate, and deploy advanced forecasting models using Python, PyTorch, and pandas.

❯❯❯❯ Deep Reinforcement Learning Hands-On: Written by Maxim Lapan, Deep Reinforcement Learning Hands-On (3rd Edition) is a detailed guide to mastering RL, covering Q-learning, DQNs, PPO, RLHF, MuZero, and transformers. With hands-on projects, it helps machine learning professionals build, train, and apply RL models using PyTorch for real-world tasks in gaming, finance, and beyond.

❯❯❯❯ Polars Cookbook: Written by Yuki Kakegawa, Polars Cookbook is a hands-on guide featuring 60+ real-world projects to master data manipulation, transformation, and analysis with Python Polars. Covering advanced querying, performance optimization, and integrations with pandas, PyArrow, and cloud platforms, this book helps data professionals build fast, scalable, and efficient workflows.

❯❯❯❯ Python Feature Engineering Cookbook: Written by Galli, Python Feature Engineering Cookbook (3rd Edition) is a practical guide featuring real-world techniques to craft powerful features for tabular, transactional, and time-series data. Covering imputation, encoding, transformation, feature extraction, and automation, this book helps data professionals build efficient, reproducible, and production-ready feature engineering pipelines.

❯❯❯❯ Data Management Strategy at Microsoft: Written by Aleksejs Plotnikovs, Data Management Strategy at Microsoft is a practical guide to building a data-driven culture and maximizing data’s business value. Covering data strategy, governance, change management, and intellectual property, it provides key insights from Microsoft’s decade-long transformation to help leaders drive impactful data initiatives.

❯❯❯❯ Zero Human Code: What I Learned from Forcing AI to Build (and Fix) Its Own Code for 27 Straight Days: This blog explores a 27-day experiment where AI tools handled all coding, debugging, and implementation while the author acted solely as an orchestrator. It reveals the real limitations of AI-driven development, challenges in guiding AI, and key insights into prompting, system complexity, and architectural rigidity.

❯❯❯❯ How LLMs Work: Pre-Training to Post-Training, Neural Networks, Hallucinations, and Inference: This blog provides a deep dive into how large language models (LLMs) work, covering their pre-training, post-training, neural network mechanics, inference, and hallucinations. It explains how LLMs are built, trained, fine-tuned, and optimized for real-world applications.

❯❯❯❯ Why Data Scientists Should Care about Containers and Stand Out with This Knowledge: This blog explains why data scientists should understand containers, particularly Docker, to enhance model deployment, reproducibility, cloud integration, and scalability. It covers key concepts, practical applications, and provides a beginner-friendly guide to setting up a Jupyter Notebook in a Docker container.

❯❯❯❯ Advanced Time Intelligence in DAX with Performance in Mind: This blog explores advanced time intelligence techniques in DAX, focusing on handling complex date-related calculations while optimizing performance. It covers scenarios like last N periods, leap years, week-to-date sums, and fiscal week YTD, using an extended date table for efficiency.

❯❯❯❯ Multimodal Search Engine Agents Powered by BLIP-2 and Gemini: This blog explores how multimodal search engine agents powered by BLIP-2 and Gemini enhance e-commerce by enabling text and image-based searches. It explains BLIP-2’s architecture, training process, and loss functions, demonstrating its application in a fashion assistant for improved product discovery.

❯❯❯❯ Formulation of Feature Circuits with Sparse Autoencoders in LLM: This blog explores how sparse autoencoders help disentangle feature circuits in large language models (LLMs), focusing on subject-verb agreement. It demonstrates how an LLM processes grammatical rules, visualizing feature circuits in both toy models and GPT-2 to enhance interpretability and debugging.

❯❯❯❯Moonshot AI Research Introduce Mixture of Block Attention (MoBA): A New AI Approach that Applies the Principles of Mixture of Experts (MoE) to the Attention Mechanism. This blog introduces Mixture of Block Attention (MoBA), a new AI approach that applies Mixture of Experts (MoE) principles to Transformer attention. MoBA improves efficiency in long-context processing by learning which token blocks to focus on, reducing computational costs while maintaining performance.

❯❯❯❯ Microsoft Researchers Present Magma: A Multimodal AI Model Integrating Vision, Language, and Action for Advanced Robotics, UI Navigation, and Intelligent Decision-Making. This blog introduces Magma, a multimodal AI model by Microsoft Research that integrates vision, language, and action for robotics, UI navigation, and intelligent decision-making. Magma outperforms existing models by combining deep learning architectures, spatial reasoning, and large-scale pretraining for superior multimodal task execution.

❯❯❯❯ Mistral AI Introduces Mistral Saba: A New Regional Language Model Designed to Excel in Arabic and South Indian-Origin Languages such as Tamil. This blog introduces Mistral Saba, a 24-billion-parameter AI model designed by Mistral AI to enhance Arabic and South Indian-origin languages like Tamil. With advanced NLP techniques and regional training, Mistral Saba delivers efficient, context-aware, and cost-effective AI solutions for diverse dialects and cultural nuances.

❯❯❯❯Publish Interactive Data Visualizations for Free with Python and Marimo: This blog explores Marimo, a newly released Python library for publishing interactive data visualizations without the need for costly servers. Combining the ease of Jupyter notebooks with Pyodide/WASM, Marimo allows data scientists to create and share interactive web-based visualizations seamlessly and for free.

❯❯❯❯ Roadmap to Becoming a Data Scientist, Part 4: Advanced Machine Learning. This blog explores advanced machine learning skills essential for data scientists, covering NLP, computer vision, reinforcement learning, and optimization techniques like fine-tuning and quantization. It emphasizes the evolution of ML methods, key concepts in LLMs, embeddings, and time series analysis, and strategies to stay competitive in the fast-changing AI landscape.

❯❯❯❯ Retrieval Augmented Generation in SQLite: This blog explores Retrieval-Augmented Generation (RAG) with SQLite, showing how to perform vector search and generative AI integration using only SQLite, the sqlite-vec extension, and OpenAI embeddings, without relying on cloud vector databases. It provides a step-by-step guide to setting up a single-file RAG system, covering virtual tables, embeddings, and querying techniques for efficient, lightweight AI applications.

❯❯❯❯ The Future of Data: How Decision Intelligence is Revolutionizing Data: This blog explores Decision Intelligence (DI), a rapidly growing field that combines AI, data science, and behavioral sciences to improve decision-making. It explains how DI differs from AI, its practical applications, and how organizations can leverage it for better predictions, automation, and efficiency across industries like retail, healthcare, finance, and manufacturing.

We’ve got more great things coming your way, see you soon!