In this Fully Connected episode, Daniel and Chris explore recent highlights from the current model proliferation wave sweeping the world - including Stable Diffusion XL, OpenChat, Zeroscope XL, and Salesforce XGen. They note the rapid rise of open models, and speculate that just as in open source software, open models will dominate the future. Such rapid advancement creates its own problems though, so they finish by itemizing concerns such as cybersecurity, workflow productivity, and impact on human culture.

Your feed might be dominated by LLMs these days, but there are some amazing things happening in computer vision that you shouldn’t ignore! In this episode, we bring you one of those amazing stories from Gabriel Ortiz, who is working with the government of Cantabria in Spain to automate cartography and apply AI to geospatial analysis. We hear about how AI tooling fits into the GIS workflow, and Gabriel shares some of his recent work (including work that can identify individual people, invasive plant species, building and more from aerial survey data).

Chris and Daniel take a step back to look at how generative AI fits into the wider landscape of ML/AI and data science. They talk through the differences in how one approaches “traditional” supervised learning and how practitioners are approaching generative AI based solutions (such as those using Midjourney or GPT family models). Finally, they talk through the risk and compliance implications of generative AI, which was in the news this week in the EU.

Our guests helped create a ML pipeline that enabled image processing and automated image comparisons, enabling healthcare use cases through their series of microservices that automatically detect, manage, and process images received from OEM equipment.

In this episode they will chat through the challenges and how they overcame them, focusing specifically on the wait strategy for their ML Pipeline Healthcare Solution microservices. We’ll also touch on how improvements were made to an open source Go package as part of this project.

There are a ton of problems around building LLM apps in production and the last mile of that problem. Travis Fischer, builder of open AI projects like @ChatGPTBot, joins us to talk through these problems (and how to overcome them). He helps us understand the hierarchy of complexity from simple prompting to augmentation, agents, and fine-tuning. Along the way we discuss the frontend developer community that is rapidly adopting AI technology via Typescript (not Python).

José Valim joins Jerod to talk all about what’s new in Livebook – the Elixir-based interactive code notebook he’s been working on the last few years.

José made a big bet when he decided to bring machine learning to Elixir. That bet is now paying off with amazing new capabilities such as building and deploying a Whisper-based chat app to Hugging Face in just 15 minutes.

José demoed that and much more during Livebook’s first-ever launch week. Let’s get into it.

Model sizes are crazy these days with billions and billions of parameters. As Mark Kurtz explains in this episode, this makes inference slow and expensive despite the fact that up to 90%+ of the parameters don’t influence the outputs at all.

Mark helps us understand all of the practicalities and progress that is being made in model optimization and CPU inference, including the increasing opportunities to run LLMs and other Generative AI models on commodity hardware.

Large Language Model (LLM) capabilities have reached new heights and are nothing short of mind-blowing! However, with so many advancements happening at once, it can be overwhelming to keep up with all the latest developments. To help us navigate through this complex terrain, we’ve invited Raj - one of the most adept at explaining State-of-the-Art (SOTA) AI in practical terms - to join us on the podcast.

Raj discusses several intriguing topics such as in-context learning, reasoning, LLM options, and related tooling. But that’s not all! We also hear from Raj about the rapidly growing data science and AI community on TikTok.

This week we’re talking about LLMs with Simon Willison. We can not avoid this topic. Last time it was Stable Diffusion breaking the internet. This time it’s LLMs breaking the internet. Large Language Models, ChatGPT, Bard, Claude, Bing, GitHub Copilot X, Cody…we cover it all.

Daniel and Chris explore the intersection of Kaggle and real-world data science in this illuminating conversation with Christof Henkel, Senior Deep Learning Data Scientist at NVIDIA and Kaggle Grandmaster. Christof offers a very lucid explanation into how participation in Kaggle can positively impact a data scientist’s skill and career aspirations. He also shared some of his insights and approach to maximizing AI productivity uses GPU-accelerated tools like RAPIDS and DALI.

Neural search and chat-based search are all the rage right now. However, You.com has been innovating in these topics long before ChatGPT. In this episode, Bryan McCann from You.com shares insights related to our mental model of Large Language Model (LLM) interactions and practical tips related to integrating LLMs into production systems.

We’ve all experienced pain moving from local development, to testing, and then on to production. This cycle can be long and tedious, especially as AI models and datasets are integrated. Modal is trying to make this loop of development as seamless as possible for AI practitioners, and their platform is pretty incredible!

Erik from Modal joins us in this episode to help us understand how we can run or deploy machine learning models, massively parallel compute jobs, task queues, web apps, and much more, without our own infrastructure.

We’re super excited to welcome Jay Alammar to the show. Jay is a well-known AI educator, applied NLP practitioner at co:here, and author of the popular blog, “The Illustrated Transformer.” In this episode, he shares his ideas on creating applied NLP solutions, working with large language models, and creating educational resources for state-of-the-art AI.

We’ve been hearing about “serverless” CPUs for some time, but it’s taken a while to get to serverless GPUs. In this episode, Erik from Banana explains why its taken so long, and he helps us understand how these new workflows are unlocking state-of-the-art AI for application developers. Forget about servers, but don’t forget to listen to this one!

Worlds are colliding! This week we join forces with the hosts of the MLOps.Community podcast to discuss all things machine learning operations. We talk about how the recent explosion of foundation models and generative models is influencing the world of MLOps, and we discuss related tooling, workflows, perceptions, etc.

Why is ML is so poorly adopted in small organizations (hint: it’s not because they don’t have enough data)? In this episode, Kirsten Lum from Storytellers shares the patterns she has seen in small orgs that lead to a successful ML practice. We discuss how the job of a ML Engineer/Data Scientist is different in that environment and how end-to-end project management is key to adoption.

Daniel and Chris do a deep dive into OpenAI’s ChatGPT, which is the first LLM to enjoy direct mass adoption by folks outside the AI world. They discuss how it works, its effect on the world, ramifications of its adoption, and what we may expect in the future as these types of models continue to evolve.

While at EMNLP 2022, Daniel got a chance to sit down with an amazing group of researchers creating NLP technology that actually works for their local language communities. Just Zwennicker (Universiteit van Amsterdam) discusses his work on a machine translation system for Sranan Tongo, a creole language that is spoken in Suriname. Andiswa Bukula (SADiLaR), Rooweither Mabuya (SADiLaR), and Bonaventure Dossou (Lanfrica, Mila) discuss their work with Masakhane to strengthen and spur NLP research in African languages, for Africans, by Africans.

The group emphasized the need for more linguistically diverse NLP systems that work in scenarios of data scarcity, non-Latin scripts, rich morphology, etc. You don’t want to miss this one!

José and Ricardo joined Daniel at EMNLP 2022 to discuss state-of-the-art machine translation, the WMT shared tasks, and quality estimation. Among other things, they talk about Unbabel’s innovations in quality estimation including COMET, a neural framework for training multilingual machine translation (MT) evaluation models.

In this special episode, we interview some of the sponsors and teams from a recent case competition organized by Purdue University, Microsoft, INFORMS, and SIL International. 170+ teams from across the US and Canada participated in the competition, which challenged students to create AI-driven systems to caption images in three languages (Thai, Kyrgyz, and Hausa).

There are some big AI-related controversies swirling, and it’s time we talk about them. A lawsuit has been filed against GitHub, Microsoft, and OpenAI related to Copilot code suggestions, and many people have been disturbed by the output of Meta AI’s Galactica model. Does Copilot violate open source licenses? Does Galactica output dangerous science-related content? In this episode, we dive into the controversies and risks, and we discuss the benefits of these technologies.

Online platforms and their users are susceptible to a barrage of threats – from disinformation to extremism to terror. Daniel and Chris chat with Matar Haller, VP of Data at ActiveFence, a leader in identifying online harm – is using a combination of AI technology and leading subject matter experts to provide Trust & Safety teams with precise, real-time data, in-depth intelligence, and automated tools to protect users and ensure safe online experiences.

Chris sits down with Ankur Goyal to talk about DocQuery, Impira’s new open source ML model. DocQuery lets you ask questions about semi-structured data (like invoices) and unstructured documents (like contracts) using Large Language Models (LLMs). Ankur illustrates many of the ways DocQuery can help people tame documents, and references Chris’s real life tasks as a non-profit director to demonstrate that DocQuery is indeed practical AI.

It’s one thing to gather some labels for your data. It’s another thing to integrate data labeling into your workflows and infrastructure in a scalable, secure, and useful way. Mark from Xelex joins us to talk through some of what he has learned after helping companies scale their data annotation efforts. We get into workflow management, labeling instructions, team dynamics, and quality assessment. This is a super practical episode!

WeightWatcher, created by Charles Martin, is an open source diagnostic tool for analyzing Neural Networks without training or even test data! Charles joins us in this episode to discuss the tool and how it fills certain gaps in current model evaluation workflows. Along the way, we discuss statistical methods from physics and a variety of practical ways to modify your training runs.