MLOps, how to build an online community, and tools for scaling LLMs: In this episode, Demetrios Brinkmann speaks to Jon Krohn about the similarities and differences between LLMOps, MLOps and DevOps, and why this should matter to companies looking to hire such engineers. You will also hear how to get involved in the MLOps community wherever you are in the world, and how you can start developing great products with the available tools.

This episode is brought to you by AWS Inferentia (https://go.aws/3zWS0au) and AWS Trainium (https://go.aws/3ycV6K0). Interested in sponsoring a SuperDataScience Podcast episode? Email natalie@superdatascience.com for sponsorship information.

In this episode you will learn:
• What MLOps is [03:51]
• About LLMOps [12:06]
• About LlamaIndex and Ollama [18:29]
• Insights from Demetrios’ MLOps survey [20:49]
• Guidance for using third-party APIs [40:18]
• Recommendations for building an online community in tech and AI [47:07]

Additional materials: www.superdatascience.com/787

The Super Data Science Podcast: Machine Learning | AI | Success

787: MLOps: The Job and The Key Tools, with Demetrios Brinkmann

Data analytics is a balance of flexibility for innovation and governance to control risks. This blog discusses its implications for artificial intelligence (AI), including machine learning (ML) and generative AI (GenAI).
Published at:
https://www.eckerson.com/articles/ai-ml-innovation-requires-a-flexible-yet-governed-data-architecture

AI/ML Innovation Requires a Flexible Yet Governed Data Architecture - Audio Blog

Dr. Julia Silge, Engineering Manager at Posit, introduces the brand-new Positron IDE, perfect for exploratory data analysis and visualization. She also lays out her top picks for LLMs that boost coding efficiency and discusses when traditional NLP methods might be the smarter choice over LLMs. Plus, Julia highlights some must-know open-source libraries that make managing MLOps easier than ever. Tune in for insights that every data scientist, ML engineer, and developer will find useful.

This episode is brought to you by Gurobi (https://www.gurobi.com/personas/optimization-for-data-scientists/), the Decision Intelligence Leader, and by ODSC (https://odsc.com/california), the Open Data Science Conference. Interested in sponsoring a SuperDataScience Podcast episode? Email natalie@superdatascience.com for sponsorship information.

In this episode you will learn:
• Overview of Posit and Positron IDE [05:20]
• How the needs of a data scientist differ from those of a software developer [10:54]
• How to contribute to the open-source Positron [19:50]
• MLOps and Vetiver: Tools for deploying and maintaining ML models [37:01]
• Natural Language Processing (NLP) and the Tidyverse approach [50:34]
• The role of AI and LLMs in data science education [1:24:18]

Additional materials: www.superdatascience.com/817

817: The Positron IDE, Tidy NLP and MLOps with Dr. Julia Silge

Welcome to "AI with Shaily"! 🎙️ I'm your host, Shailendra Kumar, and today we're diving into the exciting advancements in deep learning frameworks specifically for image recognition. So, get ready for an informative journey through the world of artificial intelligence! 🚀

First up, we have MXNet, an impressive Apache open-source framework that stands out in the fast-paced AI landscape. 🌟 This framework is like a multilingual genius, supporting programming languages such as Python, R, and C++. With its hybrid interface called Gluon, MXNet combines both imperative and symbolic programming. This unique blend allows for efficient image classification and even deepfake detection, making it a versatile tool for various programming environments—think of it as a Swiss Army knife for developers! 🛠️

Next, let's explore the realm of MLOps tools set to make waves in 2024. Picture a relay team where each member plays a critical role—this is what tools like SuperAnnotate and Encord Annotate bring to the AI landscape. 🏃‍♂️💨 These platforms specialize in image and video annotation, which enhances the accuracy of our deep learning models. Coupled with Superb AI's expertise in dataset management, we have a robust suite that ensures our data is not only accurately labeled but also efficiently processed for real-world applications. 📊

Now, imagine a hologram appearing right in front of you! ✨ That's the direction real-time image processing is heading with groundbreaking innovations like the Holoportal system. This technology captures and processes 3D video almost instantaneously, pushing the boundaries of low-latency imaging. It's not just a concept for the future; it's already transforming how we interact with visual data! 🌐

We also have to acknowledge the significant role of government initiatives in this space. The Department of Homeland Security is leveraging AI effectively, using machine learning to detect narcotics and improve investigation processes. This demonstrates the immense potential of AI in enhancing community safety and security. 🛡️

On a personal note, I’d like to share a story about a project where MXNet was pivotal in wildlife conservation. 🌿 By utilizing image recognition technology to monitor animal populations, we were able to gather data efficiently and contribute to vital ecological preservation efforts. This experience highlighted the powerful synergy between technology and nature, showcasing how AI can make a meaningful impact. 🐾

Before we wrap up, here’s a Bonus Tip for you: When embarking on image recognition projects, always consider scalability from the outset. 🏗️ MXNet’s ability to grow alongside your project's needs can save you both time and resources in the long run. 

As we conclude today’s discussion, let’s remember the insightful words of Alan Turing: "We can only see a short distance ahead, but we can see plenty there that needs to be done." With AI, our vision becomes clearer, and the possibilities are limitless! 🌈

I encourage you to follow me on YouTube, Twitter, LinkedIn, and Medium for more updates and insights. Don’t forget to subscribe to "AI with Shaily" and share your thoughts in the comments! What challenges do you think AI should tackle next? Let’s keep this conversation going! 💬

Until next time, keep questioning and keep innovating. This is Shailendra Kumar, signing off! 👋

AI Innovations Unleashed: Deep Learning Frameworks & Real-Time Image Processing

Welcome to "AI with Shaily"! 🌟 I'm Shailendra Kumar, your enthusiastic guide through the fascinating realm of artificial intelligence. Here, we explore the latest trends and insights in AI, sprinkled with my personal experiences and expert advice. 

Today, we're focusing on the essentials of MLOps best practices, specifically tailored for beginners eager to find their footing in this intricate field. Imagine stepping into a bustling engine room, where each component plays a vital role in the overall operation—this is your MLOps environment! 🛠️

We begin our journey by establishing clear paths with defined business objectives and SMART goals. Think of this as plotting a detailed roadmap for a cross-country adventure—specific destinations, measurable milestones, and time-bound checkpoints are crucial to aligning with your organizational ambitions. 🗺️

Next, we dive into the core of our engine room: data management and naming conventions. Visualize a perfectly organized library, where every book is easily accessible. This meticulous organization is essential in MLOps, especially when we consider the principle that "Changing Anything Changes Everything" (CACE). 📚

As we progress, selecting the right models and tools becomes paramount. This step is akin to assembling the ideal toolkit for our journey. Just as different terrains require various vehicles, each model is uniquely suited to our data and CI/CD pipeline. Tools like Kubernetes serve as the robust vehicles that manage our deployment journey. 🚗

Continuous monitoring acts as our reliable compass throughout the MLOps journey. Regular evaluations help us stay on course, ensuring our well-oiled machine remains efficient and effective. Methods like the ML Test Score become our guiding stars, illuminating the path ahead. 🧭

Collaboration is the heartbeat of our engine room, keeping everything in sync. Platforms like Cookiecutter data science build strong connections between teams, ensuring that data scientists, ML engineers, and DevOps experts work together seamlessly. 🤝

Finally, envision a vast reservoir capable of holding every bit of data we gather. Scalable data infrastructure is crucial for managing extensive datasets efficiently, with tools like MinIO ready to handle large volumes of data with speed and accuracy. 🌊

As we wrap up, remember this key takeaway: when mastering MLOps, always build your foundation with scalable and adaptable practices that can grow with your evolving needs. 

Before we part ways, I’d like to leave you with a thought: "The bridge between ones and zeros must be lit with collaboration, adaptability, and foresight." 🔗 Join me on social media platforms like YouTube, Twitter, LinkedIn, and Medium to share your insights, subscribe for regular AI updates, and leave your comments. Your thoughts are the fuel for our discussions. Until next time, keep your curiosity ignited and your compass pointed true! 🔥✨

Mastering MLOps: A Beginner's Guide to AI Success

PODCAST DESCRIPTION
Baselines, simple models offering reasonable results, are essential for contextualising more complex models and identifying areas for improvement. However, the sources highlight the problem of weak baselines in research, where comparisons against insufficiently tuned or considered baselines lead to misleadingly optimistic results. The authors advocate for stronger, well-tuned baselines and suggest community-level changes to improve research rigor and avoid inflated claims of progress. This includes procedural changes in conferences and a shift in research culture to prioritise quality over quantity.

DISCLAIMER
This NotebookLM-generated podcast is based on a manuscript by Dennis Li, Euxhen Hasanaj and Shuo Li, entitled "3 - Baselines". All rights are reserved to them, and uses of this reproduction should only be made for educational purposes.

Baselines In Machine Learning - A Critical Evaluation

In episode 10 of Open Source Ready, Brian and John chat with Taylor Dolezal, former CNCF Head of Ecosystem and current Chief of Staff at Merly AI, to discuss the latest trends in open source, AI, and Kubernetes. From the challenges of AI adoption to the evolving role of cloud-native technologies, Taylor shares insights on where the industry is headed and how developers and companies can navigate these rapid changes.

Subscribe

Open Source Ready - Ep. #10, The Whirlwind Pace of AI with Taylor Dolezal

This article breaks down the evolving landscape of AI/ML platforms, from AutoML to full-stack AI workbenches, and provides a structured tool evaluation framework to cut through vendor ambiguity.
Published at:
https://www.eckerson.com/articles/the-ai-ml-tool-evaluation-template-a-guide-to-smarter-selection

The AI/ML Tool Evaluation Template: A Guide to Smarter Selection - Audio Blog

Welcome to *AI with Shaily,* hosted by Shailendra Kumar, your friendly guide to the fascinating world of artificial intelligence! 🤖✨ In this episode, Shaily dives into a powerful but often overlooked innovation in machine learning called feature stores—a centralized system that organizes and manages the "features," or key data elements, that fuel ML models. Imagine it as a well-stocked pantry 🍽️ where all your essential ingredients are neatly stored and easily accessible, making your cooking (or in this case, your ML projects) smooth and efficient.

Shaily explains that feature stores have become a game-changer in MLOps (machine learning operations) as of 2025. They serve as a single source of truth, eliminating duplicated efforts and inconsistencies when different teams build models. Having worked extensively in AI development, Shaily shares a personal story where introducing a feature store transformed collaboration—teams stopped reinventing similar features under different names, saving time and boosting trust in the data. This also ensured that features used during model training matched those used during real-time prediction, leading to faster and more reliable outcomes. ⚡️🔍

Beyond organization, feature stores bring automation and pipeline management to the table. They track every change with versioning and governance—think of it like a detailed history book 📚 for your data ingredients—and perform quality checks by spotting outliers and missing values before the data reaches the model. This rigor helps build more robust and dependable ML systems.

Scalability is another highlight. Feature stores are designed to handle massive data volumes and provide real-time access, making them essential for production environments. Industry leaders like Databricks and Hopsworks offer enterprise-grade platforms, while open-source tools like Feast make these capabilities accessible to a wider audience of engineers. 🌐🚀

Shaily poses a thoughtful question: How often do ML projects slow down because teams don’t share a common “feature language”? If that sounds familiar, feature stores could be the solution you need. As a bonus tip, Shaily advises choosing feature store solutions that integrate smoothly with your existing MLOps pipeline, ensuring seamless connection between feature engineering, model training, deployment, and monitoring—key to operational success. 🔗💡

To close, Shaily quotes AI pioneer Andrew Ng: “AI is the new electricity.” Just like electricity, AI’s true power depends on solid infrastructure—and feature stores are the wiring that energizes machine learning today. ⚡️🔌

Thanks for tuning in to *AI with Shaily!* Don’t forget to subscribe on YouTube, follow Shailendra Kumar on Twitter and LinkedIn, and explore his deep-dive articles on Medium. Share your thoughts on feature stores or any AI topic in the comments—Shaily loves engaging with the community! Until next time, keep innovating and remember: the future is feature-rich! 🌟📈

Feature Stores Uncovered: The Secret to Faster, Smarter Machine Learning

Improve ML efficiency and minimize costs using ClickIT's MLOps solutions. Their team handles automation across build, test, and deployment stages, ensuring consistent performance and better stability across your entire application development lifecycle. https://www.clickittech.com/mlops-solutions/?utm_source=referral&utm_medium=backlinks&utm_campaign=rk

ClickIT’s MLOps Solutions to Optimize Processes and Drive Results

Welcome back to *AI with Shaily,* hosted by Shailendra Kumar, a lively and insightful show that brings you the latest and greatest in artificial intelligence with a personal touch! 🎙️🤖 Today’s episode dives deep into how Large Language Models (LLMs), like GPT, are revolutionizing traditional machine learning workflows in exciting new ways.

Shailendra uses a tasty cooking analogy 🍳 to explain the transformation: imagine you’ve always prepared every ingredient yourself, but now you have a smart kitchen assistant who not only helps prep but suggests new spices you never considered. That’s exactly what LLMs are doing for feature engineering — automating and enhancing the process of extracting meaningful features from complex data like text and images. This upgrade is like swapping a butter knife for a chef’s knife — sharper, more efficient, and precise. 🔪✨

The episode also highlights a hot topic buzzing on LinkedIn and Twitter: the merging of MLOps (machine learning operations) and LLMOps (operations focused on large language models). Teams are creating hybrid systems where traditional ML algorithms and LLMs coexist, deployed and monitored seamlessly. Frameworks like LangChain are gaining popularity, helping move projects from prototype to production at lightning speed. ⚙️🚀 Shailendra shares his excitement about this convergence, comparing it to his first experience automating an ML deployment — only now on a much grander scale!

Another fascinating point is how LLMs are not just analyzing data but actively augmenting the training process itself. They can generate synthetic data, provide feedback, and identify the most important samples to improve dataset quality — acting like a coach who trains and scouts the best plays on the field. 🏈📊 Plus, hybrid inference pipelines are emerging, where LLMs handle unstructured text with their natural language skills, while traditional ML methods process structured data. This combo is especially powerful in sensitive fields like healthcare and finance, where both numerical accuracy and subtle context matter deeply. 💉💰

Importantly, Shailendra emphasizes the growing trend of bringing humans back into the AI loop. This blend of AI power and human judgment ensures trustworthy, explainable outcomes. Imagine an LLM flagging a tricky prediction for a human expert’s review or generating simple explanations so users aren’t lost in technical jargon. It’s a perfect balance of machine efficiency and human insight — a crucial step toward responsible AI. 🤝🧠

Before signing off, Shailendra shares a valuable pro tip: invest time in prompt engineering tailored to your specific datasets. Fine-tuning the questions you ask LLMs can dramatically boost feature quality and overall model performance. It’s like the secret seasoning that turns a good dish into a wow-worthy meal! 🌶️✨

He leaves listeners with a thought-provoking question: How ready are you to blend these powerful language models into your existing ML setups? Are you prepared to let AI be your sous-chef in the data kitchen? 🍽️🤔

To inspire, Shailendra quotes AI pioneer Andrew Ng: “AI is the new electricity,” reminding us that this powerful force is now being wired directly into the data pipelines fueling our systems. ⚡🔌

Don’t miss out on more episodes of *AI with Shaily* available on YouTube, Twitter, LinkedIn, and Medium. Subscribe for your weekly dose of AI breakthroughs and share your thoughts to keep the conversation vibrant and exciting! 🌐📲

Till next time, keep experimenting, questioning, and AI-ing smartly! This is Shailendra Kumar signing off from *AI with Shaily.* 🎧🤓✨

LLMs as Your AI Sous-Chef: Revolutionizing Machine Learning Workflows with Hybrid Intelligence

Heavybit is the leading early-stage investor in enterprise infrastructure. Since 2013, we've helped launch and scale visionary technical startups—from DevSecOps and feature flagging to AI code generation and beyond. We go hands-on from day 0, backing founders as they turn code into companies and build the future of software from the bottom up to the top down.

Unlock the secrets of data mastery with Shailendra, a 27-year veteran in AI, ML, and Data Science, driving AU$ 6 Billion in transformative value. As the author of 'Making Money out of Data,' their strategies redefine digital success. With expertise honed at IBM and Accenture, they drive innovation and data monetization. Explore how loyalty programs like Flybuys and Everyday Rewards revolutionize business—connect with Shaily today.

The latest machine learning, A.I., and data career topics from across both academia and industry are brought to you by host Dr. Jon Krohn on the Super Data Science Podcast.

As the quantity of data on our planet doubles every couple of years and with this trend set to continue for decades to come, there's an unprecedented opportunity for you to make a meaningful impact in your lifetime. In conversation with the biggest names in the data science industry, Jon cuts through hype to fuel that professional impact.

Whether you're curious about getting started in a data career or you're a deep technical expert, whether you'd like to understand what A.I. is or you'd like to integrate more data-driven processes into your business, we have inspiring guests and lighthearted conversation for you to enjoy.

We cover tools, techniques, and implementation tricks across data collection, databases, analytics, predictive modeling, visualization, software engineering, real-world applications, commercialization, and entrepreneurship − everything you need to crush it with data science.

Listen to data and analytics leaders share the secrets of their success. Wayne Eckerson, long-time global thought leader interviews guests who run data and analytics programs at Fortune 2000 organizations around the world. Tune in to stay abreast of the latest technologies, techniques, and trends in our fast-paced industry.

Adelson de Araujo is a PhD candidate from the University of Twente. He received his MSc degree in computer science from the Federal University of Rio Grande do Norte, Brazil, where he specialized in developing AI applications for smart cities. Adelson's research projects included developing smart city applications and conversational agents for the learning sciences. In his PhD, he developed and evaluated Clair, a collaborative conversational agent to foster productive student dialogue. Adelson's recent research interests involve developing a new generation of conversational agents in education that can effectively deliver adaptive guidance to students in collaborative learning activities.

New tracks tagged #MLOps