đď¸ How I AI: How the engineer behind Claude Cowork actually uses Claude Cowork & What launched at Google I/O 2026
Your weekly listens from How I AI, part of the Lennyâs Podcast Network
How the engineer behind Claude Cowork actually uses Claude | Felix Rieseberg (Anthropic)
Listen now on YouTube ⢠Spotify ⢠Apple Podcasts
Brought to you by:
Magic PatternsâPrototypes that look like your product
GuruâThe AI layer of truth
Felix Rieseberg, the engineering lead for Claude Cowork and Claude Code Desktop at Anthropic, joins Claire to show how he actually uses Claude in his own life and work. In this episode, Felix walks through building a 3D floor planner from a 2D house plan, using email as a personal inventory database, creating live dashboards from connected apps, and hacking together a $20 hardware âClaude buddy.â He also shares his philosophy for getting more out of AI: go one abstraction layer up, let Claude work in the background, and stop assuming computers canât solve some of the annoying little problems in your life.
Biggest takeaways:
The biggest barrier to AI adoption is people not realizing they can ask AI to solve almost any problem. Felix sees this constantlyâthe tools are incredibly powerful, but users havenât built the muscle memory to reach for them. His advice: whenever youâre doing something annoying that doesnât feel creative, pause and ask yourself if Claude could do it instead. The gap isnât technical; itâs psychological.
Your email is an untapped gold mine of personal data. Felix used his email to inventory all his furniture when moving houses: every purchase receipt, every confirmation, every dimension. Claude parsed it all and built him a 3D floor planner with his actual furniture. This same principle applies to clothing, medical records, travel history, or any domain where youâve been emailing receipts and confirmations for years. You already have a structured databaseâyou just need to point Claude at it.
Go one abstraction layer up, then do it again. Felix started manually entering furniture dimensions into his floor planner, then stopped and asked Claude to figure out what furniture he had. Then he went another layer up and told Claude to find the furniture in his emails. This is the key pattern: every time you catch yourself doing tedious work, ask how Claude could do it instead. Then ask how Claude could figure out what to do without your telling it.
Live artifacts are Claudeâs answer to keeping your personal dashboards always up-to-date. Unlike static artifacts, live artifacts refresh with real-time data from your connected servicesâSpotify, Gmail, Calendar, Notion, whatever youâve authorized. Felix built a personal dashboard that looks like early-2000s software that updates throughout the day. The killer feature: you never have to manually update your pitch deck, your daily briefing, or your personal reports again.
Choose Opus when you donât know what youâre really asking for. Felixâs heuristic for model selection: use Sonnet when the problem is well-scoped and specific. Reach for Opus when you need Claude to interpret what you actually want, not just what you said. Itâs the difference between âmake me a floor plan with unitsâ (Sonnet territory) and âhelp me figure out how to organize my lifeâ (Opus territory). For most tasks, Sonnet is perfectly capable, but when you need that extra layer of problem decomposition, Opus is worth it.
Kids are the best AI users because they arenât afraid to ask for things. Felix gets videos from parents showing what their kids build with Claudeâcustom video games with hand-drawn characters, interactive stories, tools that would have required a software engineer just a few years ago. Adults have spent 20 years in a âmind prisonâ learning what computers canât do. Unlearning that is the unlock.
When Claude makes mistakes, debug your workflow, not the model. Felix doesnât curse at Claude (though he notes itâs useful for the team to know when people do). Instead, he asks it: âHereâs what I expected. Can you walk me through where things went differently? How can we prevent this in the future?â Usually the fix isnât âClaude canât do thisâ; itâs âI need to change the prompt, clean up the data source, or set up a dry run.â Treat Claude like a collaborator who needs better instructions, not a tool thatâs broken.
Blog & detailed workflow walkthroughs from this episode:
How I AI: Felix Riesebergâs Claude Workflows for 3D House Design and a $20 Hardware Buddy: https://www.chatprd.ai/how-i-ai/felix-rieseberg-claude-code-cowork-workflows-for-3d-house-design-and-hardware-buddy
âł How to Build a $20 Physical AI âBuddyâ with Claude Code: https://www.chatprd.ai/how-i-ai/workflows/how-to-build-a-20-physical-ai-buddy-with-claude-code
âł How to Create an Interactive 3D House Model from a Floor Plan Using AI: https://www.chatprd.ai/how-i-ai/workflows/how-to-create-an-interactive-3d-house-model-from-a-floor-plan-using-ai
âł How to Build a Live, Auto-Updating Personal Dashboard with Claude: https://www.chatprd.ai/how-i-ai/workflows/how-to-build-a-live-auto-updating-personal-dashboard-with-claude
What launched at Google I/O 2026 (30-minute day 1 recap)
Listen now on YouTube ⢠Spotify ⢠Apple Podcasts
Brought to you by:
Magic PatternsâPrototypes that look like your product
ThoughtSpotâBuild AI-powered analytics into your product
Claire breaks down the biggest launches from Google I/O 2026âfrom Gemini 3.5 Flash and Antigravity 2.0 to Google AI Studio, Omni, Flow, Stitch, and Pomelli. In this episode, she tests the tools live, shares what actually works, and explains where Google is catching up, where it may be pulling ahead, and why its launch-to-availability gap is still such a problem for builders.
Biggest takeaways:
Gemini 3.5 Flash rivals leading frontier coding models in Googleâs benchmarks while running four times faster. Google positions this as their agentic coding model, optimized for tasks requiring both high reasoning and rapid execution. If the benchmarks hold in practice, this speed advantage could shift the coding agent landscape toward Googleâs tools.
Antigravity 2.0 brings Googleâs IDE to feature parity with Claude Code and Codexâbut itâs playing catch-up. The update includes projects (folder-constrained workspaces), scheduled tasks on Cron, and subagents for specific tasks. The UI looks nearly identical to Codex, and the features match what Anthropic and OpenAI shipped months ago. The advantage is speed: if Gemini 3.5 Flash delivers, developers might choose Antigravity for well-scoped tasks that need to ship fast.
The /grill-me slash command is Antigravityâs aggressive take on Claude Codeâs polite clarification tool. Instead of gently asking questions, /grill-me promises to interrogate your requirements and get to the heart of what youâre building. Whether this is actually more hardcore or just clever branding remains to be seen, but it signals Googleâs attempt to differentiate on personality.
Google AI Studio now integrates directly with Workspace appsâor itâs supposed to. The promise: build no-code apps that read Sheets, draft Gmails, organize Drive, and see Calendar without setup. Claire couldnât get it to work during testing. If it delivers, it would capture internal enterprise productivity use cases and personal assistant workflows where Google already owns the data layer.
Omni is Googleâs answer to Sora, focused on longer, production-quality video. The model creates 10-second videos (versus Soraâs 6 or 7 seconds), maintains character consistency across edits, and allows conversational editing. Claire tested it by animating her kidâs drawing, and the output was impressive. The real power will be in production workflows where you iterate on the same characters and scenes multiple times.
Flow is Googleâs production-grade video editor built on Omni. It lets you define characters, create avatars, and edit videos conversationally while maintaining cinematic quality. The tool targets creators and marketers who need consistent, high-quality video at scale. Claire tried creating an avatar of herself, but the feature failedâa recurring theme throughout I/O announcements.
Stitch and Pomelli are Googleâs design and marketing tools. Stitch is like in-browser Figma with streaming design generation, inline AI edits, and code sync. Pomelli creates brand books, campaign assets, and websites from a URL. Both show promise but suffer from âGoogle slop,â the generic aesthetic of AI-generated design.
Geminiâs multimodal capabilities remain its strongest differentiator. For work involving files, videos, or transformative work across modalities (document to video, image to text), Gemini models excel. Claire uses them for generating blog posts from podcast videos and animating drawings. The 3.5 family continues this strength; for these use cases, Geminiâs multimodal performance is best-in-class.
The biggest problem: half the features donât actually work yet. Claire encountered broken features, missing integrations, and âcoming soonâ disclaimers throughout testing. Workspace integration in AI Studio? Couldnât access it. Avatar creation in Flow? Didnât work. When you announce features that arenât ready, people lose patience and stop trusting your roadmap.
Blog:
How I AI: My Live Test of Google I/Oâs New AI ToolsâFrom Gemini 3.5 Flash to Omni Video: https://www.chatprd.ai/how-i-ai/google-io-new-ai-tools-gemini-35-flash-to-omni-video
If youâre enjoying these episodes, reply and let me know what youâd love to learn more about: AI workflows, hiring, growth, product strategyâanything.
Catch you next week,
Lenny
P.S. Want every new episode delivered the moment it drops? Hit âFollowâ on your favorite podcast app.
The "go one abstraction layer up then do it again" framing is key takeaway. I keep catching myself doing the meta-work of deciding what Claude should do when the better move is just handing that question over too.