Empower Apps

Daniel Steinberg returns to chat about Apple's Foundation Models framework and what changed in OS 27. We get into using AI only where it's appropriate, experimenting with #Playground, falling back gracefully when the model isn't there, when Private Cloud Compute is worth the trip, sessions, instructions and @Generable, tools and dynamic profiles, testing with the new Evaluations framework, and why AI should learn to say "I don't know."GuestDim Sum ThinkingThe Editor's Cut - Books by Daniel SteinbergThe Puzzle of the Perfectly Paved Path: An Introduction to App IntentsTop TracksDaniel Steinberg on X (@dimsumthinking)Daniel Steinberg on Mastodon (@[email protected])Related LinksWhat's new in the Foundation Models framework (WWDC26)Build agentic app experiences (WWDC26)Build AI-powered scripts with the fm CLI and Python SDK (WWDC26)Meet the Evaluations framework (WWDC26)Build with the new Apple Foundation Model on Private Cloud Compute (WWDC26)Bring an LLM provider to the Foundation Models framework (WWDC26)Foundation Models - Apple Developer DocumentationImage Playground - Apple Developer DocumentationShip the Fallback First - Jeroen LeenartsDo iOS 2026 - Intro to Foundation Models workshopBushel - macOS Virtual Machine App for DevelopersRelated EpisodesFunctional Programming with Daniel SteinbergDo the Wrong Thing Much Faster with Graham LeeWe Don't Have Any Futures or Promises Anymore with Tim CondonPractical Agents with Donny WalsEveryone Thinks They're Good at Prompting with Joe FabisevichChapters(00:00) - Stage Three: Using AI Where Appropriate (05:19) - Experimenting with #Playground (07:36) - What's New in OS 27 (13:33) - Fallbacks and Private Cloud Compute (21:12) - Sessions, Instructions, and @Generable (28:26) - Tools, Dynamic Profiles, and Streaming (37:12) - Images and Testing with Evaluations (43:21) - Platforms, Privacy, and the Cost of Tokens WatchClick here to watch a video of this episode. TranscriptClick here to view the episode transcript. Support the Show ★ Support this podcast on Patreon ★  Thanks to our supporters: Thanks to our monthly supporters work design Steven Lipton  Welcome new supporters: Social MediaLinkedIn - @leogdionGitHub - @brightdigitGitHub - @leogdionMastodon - @[email protected] - @brightdigitX - @leogdionX - @brightdigitCreditsMusic from https://filmmusic.io "Blippy Trance" by Kevin MacLeod (https://incompetech.com) License: CC BY (http://creativecommons.org/licenses/by/4.0/)

Graham Lee returns to chat about what it takes to become a software centaur: using AI to get better at the craft, not just faster. We get into why the job was never about typing code, meat proxies and Jevons paradox, his catalog of interaction patterns for AI-augmented development, why a model shouldn't mark its own homework, and Axminster, his UML-based macOS app that gives LLMs a design to follow. GuestChiron CodexChiron Codex PatreonChiron Codex Agent Skills - GitHubStructure and Interpretation of Computer Programmers podcastSoftware Craftsmanship's Consulting DetectiveRelated LinksAxminster: AI-Augmented Software DesignAxminister - Mac App StorePatterns in AI Augmented Software Development - iOSDevUKSoftware Centaurs white paper (PDF)Reasons to dislike AI codingReplacing MCP with Standard Filesystem ToolsIs spec-driven development the end of agile?Episode 64: Using Pattern Languages for Object-Oriented ProgramsObject-Oriented Programming the Easy WayRelated EpisodesThe More You Know with Stewart LynchSkynet Came and Automated Everything with Jared SorgePractical Agents with Donny WalsEveryone Thinks They're Good at Prompting with Joe FabisevichActually Really UsefulObjective-C Today with Graham LeeChaptersWatchClick here to watch a video of this episode. TranscriptClick here to view the episode transcript. Support the Show ★ Support this podcast on Patreon ★  Thanks to our supporters: Thanks to our monthly supporters work design Steven Lipton  Welcome new supporters: Social MediaLinkedIn - @leogdionGitHub - @brightdigitGitHub - @leogdionMastodon - @[email protected] - @brightdigitX - @leogdionX - @brightdigitCreditsMusic from https://filmmusic.io "Blippy Trance" by Kevin MacLeod (https://incompetech.com) License: CC BY (http://creativecommons.org/licenses/by/4.0/)

Georg Tuparev stops by to chat about software that outlives its plans: projects scoped for a few months that are still in development 20 years later. We get into running a small team of physicists and astronomers, why WebObjects and EOF are still in production, where AI helps and where it falls short, server-side Swift with SwiftNIO, and storing exabytes of telescope data from observatories no cable can keep up with.GuestGeorg Tuparev - Tuparev Technologies@[email protected] - MastodonGeorg Tuparev - LinkedIntuparev (Georg Tuparev) - GitHubGeorg Tuparev (@tuparev) - XSkyObserverRelated LinksExabyte Data Storage - Georg Tuparev - ServerSide.swiftExabyte Data Infrastructure - Tuparev TechnologiesAstronomy & Space - Tuparev TechnologiesSwiftNIO - GitHubServerSide.swiftRelated EpisodesWe Don't Have Any Futures or Promises Anymore with Tim CondonThe More You Know with Stewart LynchFull Stack Things with Werner Jainek and Vojtěch RylkoObjective-C Today with Graham LeeChaptersWatchClick here to watch a video of this episode. TranscriptClick here to view the episode transcript. Support the Show ★ Support this podcast on Patreon ★  Thanks to our supporters: Thanks to our monthly supporters work design Steven Lipton  Welcome new supporters: Social MediaLinkedIn - @leogdionGitHub - @brightdigitGitHub - @leogdionMastodon - @[email protected] - @brightdigitX - @leogdionX - @brightdigitCreditsMusic from https://filmmusic.io "Blippy Trance" by Kevin MacLeod (https://incompetech.com) License: CC BY (http://creativecommons.org/licenses/by/4.0/)

Tim Condon stops by to chat about Vapor 5, the biggest Vapor release in six and a half years. We get into dropping futures and promises for Swift concurrency all the way down the stack, Span and strict memory safety, and the new shared Swift HTTP server.GuestTim CondonTim Condon (@[email protected]) - MastodonTim Condon (@0xtim.bsky.social) - Bluesky0xTim (Tim Condon)VaporRelated LinksVapor 5 Beta - Vapor BlogWhat's New in Vapor 5 Beta - Vapor BlogAnnouncing the Networking Workgroup - Swift.orgswift-http-server - GitHubSwift OpenAPI Generator - GitHubRelated EpisodesYear of the Server with Tim Condon - Part 1Year of the Server with Tim Condon - Part 2Server-Side Swift with Tim CondonWho's Wendy with Joannis OrlandosSwift Server Workgroup with Joannis OrlandosChapters(00:00) - Server-Side Swift Today (02:29) - Vapor 5, Span & Memory Safety (13:38) - Package Flags & Dropping EventLoopFutures (18:36) - Streaming & the Swift HTTP Server (24:21) - VaporTesting & Macros (33:18) - OpenAPI & the Future of Fluent (38:48) - AI, Open Source & the New Networking APIs WatchClick here to watch a video of this episode. TranscriptClick here to view the episode transcript. Support the Show ★ Support this podcast on Patreon ★  Thanks to our supporters: Thanks to our monthly supporters work design Steven Lipton  Welcome new supporters: Social MediaLinkedIn - @leogdionGitHub - @brightdigitGitHub - @leogdionMastodon - @[email protected] - @brightdigitX - @leogdionX - @brightdigitCreditsMusic from https://filmmusic.io "Blippy Trance" by Kevin MacLeod (https://incompetech.com) License: CC BY (http://creativecommons.org/licenses/by/4.0/)

Stewart Lynch stops by to chat about using AI to learn faster instead of just letting it write your code. We get into planning in stages before you build, why Markdown is the universal language of AI, and the tiny "donkey work" tools he vibe codes for himself.GuestStewart LynchStewart Lynch 🇨🇦 (@[email protected]) - MastodonStewart Lynch (@stewartlynch.bsky.social) - BlueskyStewart Lynch - YouTubeStewart Lynch | LinkedInStewart Lynch - All Social LinksCreatech Solutions BlogRelated LinksHow AI assistance impacts the formation of coding skills - AnthropicSosumi.ai - Apple docs as MarkdownTypora - Markdown editorRelated EpisodesHow to Learn New APIs with Stewart Lynch - Part 1How to Learn New APIs with Stewart Lynch - Part 2Practical Agents with Donny WalsEveryone Thinks They're Good at Prompting with Joe FabisevichChapters(00:00) - Xcode 27 & the New Device Hub (03:36) - Why YouTube Teaching Is Changing & Using AI to Synthesize New APIs (11:24) - Why Learn When AI Can Build It? Staged Plans & PRDs (16:29) - Markdown as the Universal Language of AI (22:07) - AI Costs, Hallucinations & Learning Smarter (27:48) - Xcode + Claude Code Workflow (30:53) - "Donkey Work": Building Small Tools With AI (35:55) - What to Delegate to AI vs. Control Yourself (40:22) - OS 27 Support, the iPhone Dual & Side-Project Playgrounds WatchClick here to watch a video of this episode. TranscriptClick here to view the episode transcript. Support the Show ★ Support this podcast on Patreon ★  Thanks to our supporters: Thanks to our monthly supporters work design Steven Lipton  Welcome new supporters: Social MediaLinkedIn - @leogdionGitHub - @brightdigitGitHub - @leogdionMastodon - @[email protected] - @brightdigitX - @leogdionX - @brightdigitCreditsMusic from https://filmmusic.io "Blippy Trance" by Kevin MacLeod (https://incompetech.com) License: CC BY (http://creativecommons.org/licenses/by/4.0/)