How I AI
Claire Vo
How I AI, hosted by Claire Vo, is for anyone wondering how to actually use these magical new tools to improve the quality and efficiency of their work. In each episode, guests will share a specific, practical, and impactful way they’ve learned to use AI in their work or life. Expect 30-minute episodes, live screen sharing, and tips/tricks/workflows you can copy immediately. If you want to demystify AI and learn the skills you need to thrive in this new world, this podcast is for you.
Episodes
Mentioned books
98 snips
Oct 14, 2026 • 24min
Meta’s new ‘Muse Wonder’ model: good for AI agents, bad for design
Meta’s Muse Wonder models face nearly 20 tests spanning coding, research, email, voice assistants, agent workflows, SVG illustration, and short-form video. The results reveal standout performance in complex agentic tasks and backend implementation, while design and creative output fall noticeably behind. A billing workflow puts its sustained focus to the test, and comparisons with leading models highlight where Muse belongs in real-world workflows.
359 snips
Oct 12, 2026 • 38min
How Meta designed Muse: an AI agent people actually want to talk to | Alex Cornell (Design Director)
Alex Cornell, Meta design director leading Muse, explores how to make AI agents feel natural and approachable. He discusses messaging-inspired interfaces, layered status updates, generative UI, and turning Instagram saves into a personalized shopping catalog. He also shares creative uses for golf tracking, archived blogs, and chess analysis, plus the design challenges of unpredictable AI outputs.
363 snips
Oct 5, 2026 • 36min
How OpenAI uses ChatGPT Sites (live at DevDay!) | Kath Korevec (Product Lead)
Kath Korevec, an OpenAI product lead working on Codex and ChatGPT Sites, showcases software built with connected data and plugins. She explores an incident command site powered by Slack, Notion, and calendars, a family-proof Spotify workflow, and a community dungeon crawler. The conversation also covers creative momentum, fast models, shared AI-built worlds, and why she won’t let AI speak in her name.
388 snips
Sep 30, 2026 • 46min
Jev: 8 real use cases for the fastest, cheapest model I’ve ever used | John Lindquist
John Lindquist, creator of developer education platform egghead.io and founder of mega.dev, explores Jev, a fast, low-cost decision model. He demos voice-powered to-do automation, instant data deduplication, natural-language app routing, chess comparisons, browser navigation, multi-agent coordination, and a real-time presentation coach. They also discuss multi-pass classification and when traditional generative models remain the better choice.
258 snips
Sep 30, 2026 • 24min
OpenAI Dev Day 2026: The releases that actually matter
OpenAI’s latest developer releases get a hands-on test drive. Explore persistent Dots, human-agent collaboration in Spaces, governed internal tools through Sites, and the fast GPT-6.1 Sol model. See vision-powered thumbnail decisions, a collaborative AI sketchpad, and a real-time 3D world redesigned by kids, with one experiment running up a $97 bill.
432 snips
Sep 28, 2026 • 26min
Jev for beginners: how to use it and what to build
Jev turns AI into a fast decision engine for choices, scores, and probabilities. See how it categorizes 1,700 pull requests, analyzes coding sessions, filters Gmail, organizes product signals, and transforms 4,500 YouTube comments into a searchable audience dashboard. Also featured: a real-time voice-to-color emotion app and creative ways to pair Jev with larger models.
332 snips
Sep 22, 2026 • 39min
Opus 5.5 vs. GPT-6 Sol: which model won my blind taste test?
A live blind taste test pits Opus 5.5, GPT-6 Sol, GPT-6 Astra, and more against emails, PRDs, frontend and backend work, long-running agents, SVGs, and video editing. The results spark surprises, shifting predictions, an LLM judge disagreement, and a return to the notoriously difficult Barbie 3D benchmark, where better rendering still cannot rescue the hands.
320 snips
Sep 22, 2026 • 25min
I left Claude for months. Opus 5.5 is why I'm back
After months away from Claude, Claire Vo puts Opus 5.5 through demanding real-world tests, from long-running agents and inbox triage to a ChatPRD homepage redesign, SVG illustration, voice assistance, and video editing. She explores its speed, pricing, safety boundaries, prompt-injection resistance, and surprising strengths, while comparing it with Codex and examining where it still falls short.
307 snips
Sep 21, 2026 • 47min
How Warp ships 2,000 PRs a month with AI factories | Zach Lloyd (CEO, Warp)
Zach Lloyd, co-founder and CEO of Warp and former Google principal engineer, explores AI software factories that turn Slack ideas into merged pull requests. He discusses measuring human interactions, reshaping code review, scoring agent runs, learning from failures, and balancing model cost with quality. Zach also demos AI-powered workflows for Figma, meeting notes, lead research, and CEO productivity.
494 snips
Sep 16, 2026 • 37min
Muse review: The personal AI agent that gets consumer UX right
A hands-on tour of Meta’s Muse explores family calendar management, personalized morning newsletters, wellness goals, media-rich libraries, and browser-based errands. The review digs into transparent activity tracking, progressive permissions, and an unexpectedly delightful animated avatar. Shopping tests reveal both impressive successes and frustrating limits, offering a vivid look at where personal AI agents shine today.
