Changelog Master Feed

Chip Huyen explores the shifting sand of large data models, Herman Õunapuu reviews the Zimaboard, Bryan Braun shares recent VSCode configuration discoveries, and Swizec Teller summarizes the inaugural AI Engineer Summit.

In this episode, the hosts discuss their recent experiences with various improvements including enabling HTTP/3, Brotli compression, and Fastly websockets. They also talk about the importance of continuous improvement and the consequences of ignoring small problems at scale. The discussion covers advancements in data and machine learning, the challenges of enabling WebSockets for O-BAN Web, and the impact of sophisticated tooling on engineers. They also reflect on attending a large conference, deliberate networking, and share their contact information for further conversations.

Sunil Pai, founder and CEO of PartyKit, joins the show to talk about PartyKit, an open-source tool for creating collaborative, multiplayer applications. They discuss the evolution of real-time technology, the features and benefits of PartyKit, and the landscape of collaborative tools. They also explore building on cloud infrastructure for scalability and the structure and roles of the founder team. A fun and informative episode!

A lightweight, text-based terminal recording project called asciinema, its origins in the 'script' command, challenges encountered in recording terminals on different computers, and its potential for teaching and documentation purposes. The features of asciinema such as manipulating terminal events and changing the theme, its financial stability, and future prospects. The decision to keep the project free and open, avenues for sustainability, and securing monthly sponsors. The idea of adding audio to tutorials and the possibility of incorporating it into the text-based format. Exporting and using the command line tool, including exporting locally stored or website-built ones and using Ag for social media preview cards and gif generation. The upcoming online conference 'nodes' for developers and data scientists to learn about graph technology. Features they want to implement, their sponsors, and the community gathering in their matrix room. Bonus content for subscribers and the upcoming Kizen edition of the podcast.

Angelica, Johnny, Kaylyn, and Kris discuss their experiences at GopherCon 2023, including organizing the conference, live captioning, meetups, opinions on JavaScript vs. Go, balancing data-driven decisions, and lighthearted conversations.

Jacob Kaplan-Moss discusses remote and colocated teams. Duarte Carmo created a neural search engine from Changelog transcripts. Tom Hacohen advocates for strong static typing. Orhun Parmaksız created a CLI to make your keyboard sound like a typewriter. Luke Plant shares hard truths about simplicity.

Guest Christina Warren from GitHub joins the hosts to discuss the end of physical media, reminiscing about DVDs and Blu-rays. They talk about the decline of physical media, the role of Redbox, and the convenience of digital formats. They also explore the impact of technology on the film industry and express concern over the pressure for every film to be a blockbuster. Reflecting on the transition to digital formats, they bid farewell to physical media.

Jarred Sumner, creator of Bun, an all-in-one JavaScript runtime, reflects on Bun's big 1.0 launch. Topics discussed include problem-solving, speed, no Windows support, critics, Bun vs. Node, sustaining the startup, and more.

Daniel Thompson, Co-founder and Core Member of Tauri, talks about the updates on Tauri 1.5, the launch of Crab Nebula, utilizing Tauri ecosystem, comparing Electron and Tauri for user interface, the growing adoption of Tauri, building a product and dev tools, and the features of an ideal web browser.

Chris and Daniel discuss the model lifecycle for generative AI models, including optimization and serving. They explore deploying models, finding open access models, and navigating the landscape of generative models. They also touch on strategies for optimizing and deploying AI models, and different ways of deploying generative models.