You might have heard that “AI is only as good as the data.” What does that mean and what data are we talking about? Chris and Daniel dig into that topic in the episode exploring the categories of data that you might encounter working in AI (for training, testing, fine-tuning, benchmarks, etc.). They also discuss the latest developments in AI regulation with the EU’s AI Act coming into force.

The incomparable Jessica Kerr drops by with a grab-bag of amazing topics. Understanding software systems, transferring knowledge between devs, building relationships, using VS Code & Docker to code together, observability as a logical extension of TDD, and a whole lot more.

Google has been releasing a ton of new GenAI functionality under the name “Gemini”, and they’ve officially rebranded Bard as Gemini. We take some time to talk through Gemini compared with offerings from OpenAI, Anthropic, Cohere, etc.

We also discuss the recent FCC decision to ban the use of AI voices in robocalls and what the decision might mean for government involvement in AI in 2024.

This week we’re joined by long-time web developer Matt Patterson. Earlier this year Matt wrote an evocative article for A List Apart called The Future of Web Software Is HTML-over-WebSockets. In this episode Matt sits down with Jerod to discuss, in-detail, why he believes the future of the web is server-rendered (again) and how Ruby on Rails is well positioned to bring that future to us today.

This week we’re taking you to the hallway track of All Things Open 2022 in Raleigh, NC. Let’s set the stage, here’s what we like do when we go to conferences — we setup our podcast studio at our booth where all the other vendors are and we talk to everyone we can. We give out t-shirts, stickers, pins, high fives…and it’s a blast.

Today’s anthology episode from ATO features: Arun Gupta (VP and GM of Open Ecosystem Initiatives at Intel), long-time friend Chad Whitacre (Head of Open Source at Sentry), and Ricardo Sueiras (Principal Advocate in Open Source at AWS).

The common denominator for each of these conversations is advocating for and supporting open source. Special thanks to Todd Lewis and team for inviting us to come back to ATO. We enjoyed meeting long time fans and new ones too.

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.

Simon Willison rounds up the goings on around Microsoft’s new GPT-powered Bing search, The Vue/Vite team build a nimble web client for Mastodon, Will Larson writes about writing an engineering strategy, Denis Pushkarev seeks support to maintain core-js & I share a lightning round of cool tools I’ve found and used recently. ⚡️

This week Daniel & Chris hang with repeat guest and good friend Demetrios Brinkmann of the MLOps Community. Together they review, debate, and poke fun at the 2024 Gartner Hype Cycle chart for Artificial Intelligence. You are invited to join them in this light-hearted fun conversation about the state of hype in artificial intelligence.

This week we’re talking to Daniel J. Barrett, author of Efficient Linux at the Command Line as well as many other books. Daniel has a PhD and has been teaching and writing about Linux for more than 30 years (almost 40!). So we invited Dan to join us on the show to talk about efficient ways to use Linux. He teaches us about combining commands, re-running commands, $CDPATH hacks, and more.

How do you systematically measure, optimize, and improve the performance of LLM applications (like those powered by RAG or tool use)? Ragas is an open source effort that has been trying to answer this question comprehensively, and they are promoting a “Metrics Driven Development” approach. Shahul from Ragas joins us to discuss Ragas in this episode, and we dig into specific metrics, the difference between benchmarking models and evaluating LLM apps, generating synthetic test data and more.

The new open source AI book from PremAI starts with “As a data scientist/ML engineer/developer with a 9 to 5 job, it’s difficult to keep track of all the innovations.” We couldn’t agree more, and we are so happy that this week’s guest Casper (among other contributors) have created this resource for practitioners.

During the episode, we cover the key categories to think about as you try to navigate the open source AI ecosystem, and Casper gives his thoughts on fine-tuning, vector DBs & more.

WestArtFactory’s premium PCB cheat sheets, Maxime Topolov tells of a disappearing AWS dev, Juspay Technologies releases HyperSwitch for payment processing, Servo gets new funding for 2023 & Cloudflare’s open source Wildebeest.

Erez Zukerman shares the story of launching the ErgoDox EZ on Indiegogo (May 2015), what it takes to create customizable ergonomic keyboards, the benefits of split keyboards and custom key layouts, repairability and longevity, community engagement, and the attention to detail required in everything they create. We talk through their keyboard lineup, our personal experience with how we mouse and keyboard…we cover it all.

Reorx lists awesome apps & tools using the new ChatGPT API, Ernie Smith ranks self-hosted app alternatives, Very Good Ventures brings Dart to the server, Daniel Stenberg tells curl’s NuGet story & Hacker Stations showcases tech workspace setups from all over the world.

Workflow orchestration has always been a pain for data scientists, but this is exacerbated in these AI hype days by agentic workflows executing arbitrary (not pre-defined) workflows with a variety of failure modes. Adam from Prefect joins us to talk through their open source Python library for orchestration and visibility into python-based pipelines. Along the way, he introduces us to things like Marvin, their AI engineering framework, and ControlFlow, their agent workflow system.

Ellie Huxtable’s Atuin makes your shell history magical, Dmitry Kudryavtsev writes why he thinks engineers should focus on writing, LazyVim promises to transform your Neovim setup into a full-fleged IDE, Geoff Graham shares with Smashing Magazine how he writes CSS in 2023 & Brad Fitzpatrick collects a public list of bad issue track behaviors.

Event-driven systems may not be the go-to solution for everyone because of the challenges they can add. While the system reacting to events published in other parts of the system seem elegant, some of the complexities they bring can be challenging. However, they do offer durability, autonomy & flexibility.

In this episode, we’ll define event-driven architecture, discuss the problems it solves, challenges it poses & potential solutions.

Jerod & Adam share our Zulip first impressions, react to Elasticsearch going open source (again), discuss Christian Hollinger’s blog post on why he still self-hosts & answer a listener question: how do we produce podcasts?

Chris & Daniel are out this week, so we’re bringing you a panel discussion from All Things Open 2023 moderated by Jerod Santo (Practical AI producer and co-host of The Changelog) and featuring keynoters Emily Freeman and James Q Quick.

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.

Jerod and Adam talked with Rich Harris –a JavaScript Journalist on The New York Times Investigations team– about his magical disappearing UI framework called Svelte. We compare and contrast Svelte to React, how the framework is embedded in a component, build time vs. run time, scoping CSS to components, and CSS in JavaScript. Rich also shares where Svelte v3 is heading and the details on Sapper, a framework for building extremely high-performance progressive web apps, powered by Svelte.

It seems like everyone is interested in Rust these days. Even the most popular Python linter, Ruff, isn’t written in Python! It’s written in Rust. But what is the state of training or inferencing deep learning models in Rust? In this episode, we are joined by Nathaniel Simard, the creator burn. We discuss Rust in general, the need to have support for AI in multiple languages, and the current state of doing “AI things” in Rust.

Jerod is joined by Chris and Desmond (co-hosts of the ElixirTalk podcast) to catch up on what’s moving and shaking in the Elixir and Phoenix communities. We discuss what’s attractive about Elixir, what it means to have the language finalized, why folks are so excited by Phoenix LiveView, the ambitious new Lumen project that’s bringing Elixir to WebAssembly, and more.

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).

Sahn Lam details Stack Overflow’s monolith/on-prem architecture, Hillel Wayne asks the Lobsters community for killer libraries, Linux 6.2 is ready to run on M1 Macs thanks to Asahi Linux, Johan Halse writes up what to expect from your web framework & Eli Bendersky on using GoatCounter for blog analytics.

Ken Thompson’s 75-year-project is a jukebox for the ages, Tabby is a self-hosted AI coding assistant, Codeberg is a collaboration platform and Git hosting for open source software, content and projects, TheSequence explains The LLama Effect & Paul Orlando writes about Ghosts, Guilds and Generative AI.

This week Adam talks with Marcin Kurc about chasing the 9s. Marcin is the Co-founder and CEO of Nobl9 where they build tools for managing service level objectives, aka SLOs. We also talk about service level agreements (SLAs), service level indicators (SLIs), error budgets, and monitoring, and how it all comes together to help teams align on goals, improve customer satisfaction, manage risks, increase transparency, and of course, a favorite around here…continuous improvement. Kaizen! This is an awesome deep dive into the world of chasing those 9s, and how teams are levering SLOs to earn the trust of their customers as well showcase transparency.

Emily Freeman joins the show alongside our Ship It co-host, Justin Garrison! We hear Emily’s burnout story & learn how she and Forrest Brazeal are putting tech-focused influencers on tap. But first: area code turf wars, bad movie reboots & buying used DVDs… at Starbucks?!

Daniel & Chris engage in an impromptu discussion of the state of AI in the enterprise. Then they dive into the recent Apple Intelligence announcement to explore its implications. Finally, Daniel leads a deep dive into a new topic - Advanced RAG - covering everything you need to know to be practical & productive.

We’re excited to have Tuhin join us on the show once again to talk about self-hosting open access models. Tuhin’s company Baseten specializes in model deployment and monitoring at any scale, and it was a privilege to talk with him about the trends he is seeing in both tooling and usage of open access models. We were able to touch on the common use cases for integrating self-hosted models and how the boom in generative AI has influenced that ecosystem.