Peter Yang

Jared Zoneraich (Builder-in-Residence at Cognition) shares his most radical prediction for how we'll work with AI agents. "I think humans are going to return to a world where we just have a wooden desk — no monitor setup, no keyboard, no mouse. I just want a button and a whisper flow. And that's all I need." 📌 Watch the full conversation below. 🎧 Full episode: https://youtu.be/0YeeJHYy-Vc #AI #AIagents #futureofwork #Cognition #Shorts

Jared Zoneraich (Builder-in-Residence at Cognition) on the #1 mistake teams make when building AI agents. "You want to build in a way that lines up with the roadmap of the model labs. Let the model cook, and don't try to push it in certain directions. You'll need less scaffolding around agents as the models get better. And they're pretty good already." 📌 Watch the full conversation below. 🎧 Full episode: https://youtu.be/0YeeJHYy-Vc #AI #AIagents #coding #Cognition #Shorts

Jared Zoneraich (Builder-in-Residence at Cognition) on agent fan-out and why the human shouldn't be the bottleneck anymore. "More Devins are being launched by other Devins than by humans. Every time a Datadog alert goes off, you can pipe it to Devin. Like the early days of self-driving cars — you didn't want the GPU sitting idle. If I'm going to sleep, I want my agents doing stuff." 📌 Watch the full conversation below. 🎧 Full episode: https://youtu.be/0YeeJHYy-Vc #AI #AIagents #coding #Devin #Cognition #Shorts

Jared is a Builder in Residence at Cognition, the team behind Devin. He showed me why teams often overbuild their first agents, and how simple prompts, better tools, and self-testing work together instead. Then he demoed one manager agent breaking down a task, directing 10 cloud agents, and deciding what a human needs to review. Jared and I talked about: (00:00) Why teams overbuild their first AI agent (02:50) Let the model cook before adding rules (03:16) Why tools matter more than better prompts (10:51) Why you should build evals after shipping (16:06) What cloud agents can do that local agents can’t (20:30) How to get one agent to manage other agents (25:39) Demo: One agent launches a team of 10 (28:03) How agents prove their work before you review it (33:18) The best tasks for agent teams to work on Thanks to our sponsors: Riverside: All-in-one AI studio for podcasts and video https://creators.riverside.com/PeterYang Linear: The AI agent platform for modern teams https://linear.app/behind-the-craft Wispr Flow: 4x faster than typing with your voice https://ref.wisprflow.ai/peteryang 📌 Get the takeaways: https://creatoreconomy.so/p/how-to-build-ai-agents-that-check-their-own-work-jared-cognition-devin 📌 Get my personal AI OS with useful AI skills, $270 in AI tool credits, live monthly workshops, and courses on Hermes and Codex agents: https://www.behindthecraft.com/ Where to find Jared: X: https://x.com/imjaredz Devin: https://devin.ai Subscribe to this channel - more interviews coming soon!

GPT-5.6 burns through more tokens than GPT-5.5. My recommendation for everyday work: "Use GPT-5.6 Sol at medium effort. High effort or the other models burn tokens super fast. Sol at medium is a great everyday workhorse for OpenClaw, Hermes, browser tasks, and everything else." 📌 Visit my YouTube for more videos.

OpenAI just made GPT 5.6 available to everyone and I’m going to answer your most burning question: Is GPT 5.6 better than Claude Fable 5? Watch my tutorial to see me test both models head-to-head across 6 real use cases. Timestamps (00:00) GPT-5.6 vs Claude Fable 5 (00:46) Compare model tiers, price, and coding performance (01:45) Create an interactive travel website (05:44) Build a 3D Star Fox-style game in one shot (09:39) Edit and publish short videos with browser use (15:10) Plan and build a mobile app feature (19:34) Get personalized life and business advice (22:17) Improve and clean up a personal AI OS (24:37) Comparing each model’s strengths and weaknesses (26:36) My daily-driver recommendation Get the written post: https://creatoreconomy.so/p/gpt-vs-claude-fable-i-tested-6-real-use-cases Get my personal AI operating system with 17+ AI skills, full courses, and more: https://www.behindthecraft.com/ Where to find me: Newsletter: https://creatoreconomy.so/ X: https://x.com/petergyang LinkedIn: https://www.linkedin.com/in/petergyang/ Subscribe to this channel - more tutorials coming soon!

Codex isn’t just for shipping code. It can also help protect your calendar. From Rohan (Codex PM): “I probably have a pretty large number of meetings relative to what would be optimal, but a lot of that’s actually just with customers. I would say my build time is from 5 PM onwards. But it’s funny because I’ll even use Codex as my personal chief of staff. At the beginning of the week, I’ll have it look at all my meetings, consolidate the ones where it can give me focus blocks, and figure out which ones could be canceled as a DM or something. So it’s definitely helping me find the focus time. But in general, what we do with the free time is actually just spend way more time with our customers.” 📌 Watch the full conversation below.

“Developers even haven’t fully warmed up to the level of ambition you can have with Codex." From Rohan (Codex PM): "If you go into any large enterprise, maybe the top five to ten percent most AI-pilled frontier engineers are maximally leveraging Codex. But then there’s a long tail of people that are still maybe pair programming, doing less delegation. Engineers have seen AI incrementally expand its capabilities over the last three years. But if you look to a non-developer, it’s kind of been just ChatGPT this whole time. The type of things they’re asking ChatGPT are very different than what you can actually delegate to Codex. Almost everything you need to do professionally should almost always just immediately have Codex start doing it.” 📌 Watch the full conversation below.

"Codex is the only way I keep track of everything." From Rohan (Codex PM): "Whenever I’m in meetings, or literally anytime I don’t understand something or hear something I didn’t know about, I just have Codex go and gather all the context. A couple times a day there’s something someone is doing where I’m like, it’s incredible that they’re doing it. I had no idea it was happening. You cannot walk from point A to point B in the OpenAI office without hearing Codex get mentioned.” 📌 Watch the full conversation below.

As a product manager for OpenAI Codex, Rohan has a firsthand look at how one of the most AI-native teams in the world operates. He showed me how he uses Codex to trigger work from Slack replies, prototype designs with image generation, replace stale PRDs with live project sites, and more. Watch this episode if you’re tired of all the nonsense around “AI-native PM” and want to see how a real one actually operates. Rohan and I talked about: (00:00) Codex can do more than you think (01:53) Use Codex to pull context from Slack, docs, Linear, and email (03:28) How Codex flips planning from docs to demos (13:14) Demo: Do the task once with Codex, then automate it (14:30) Demo: Trigger Codex automations from Slack replies (17:26) Demo: Use Image Gen to explore design variants fast (19:58) Demo: Turn any Codex thread into a reusable skill (29:55) Demo: Turn project context into a shareable Site (33:33) Use one Codex thread to manage other Codex threads (37:18) Aim 10x more ambitious than feels reasonable Thanks to our sponsors: Linear: The AI agent platform for modern teams https://linear.app/behind-the-craft Wispr Flow: 4x faster than typing with your voice https://ref.wisprflow.ai/peteryang Riverside: All-in-one AI studio for podcasts and video https://creators.riverside.com/PeterYang 📌 Get the takeaways: https://creatoreconomy.so/p/openai-pm-reveals-how-he-uses-codex-for-product-work-rohan-varma 📌 Get my personal AI OS with useful AI skills, $270 in AI tool credits, live monthly workshops, and courses on Hermes and Codex agents: https://www.behindthecraft.com/ Where to find Rohan: LinkedIn: https://www.linkedin.com/in/rohannvarma X: https://x.com/TheRohanVarma Codex: https://openai.com/codex/ Subscribe to this channel - more interviews coming soon!

Most apps are still designed for humans clicking buttons. But AI agents will increasingly read, write, buy, browse, and operate software on our behalf. If agents can work across many apps, a single app’s chatbot won’t be enough. 📌 Visit my YouTube for more videos.

Gemini Spark is Google's 24/7 personal AI agent. In this tutorial, I show 5 practical use cases for Spark across Gmail, Calendar, Google Flights, Maps, Docs, and Sheets, plus my honest take on where Spark is useful and where it still falls short. What I cover: (00:00) Why Gemini Spark is the most beginner-friendly agent (01:47) How to switch from Gemini chat to Spark (02:12) Triage urgent emails with Spark and Gmail (04:53) Prep for meetings with Spark and Google Calendar (08:33) Plan trips with Spark and Flights, Maps, and Docs (13:28) Track summer camp deadlines with Spark and Sheets (15:59) Two useful Spark skills you should try (19:30) Where Spark is useful, and where it falls short This video is sponsored by Google but opinions are my own. Try Spark here: https://gemini.google.com/spark Get my personal AI operating system with 17+ AI skills, full courses, and more: https://www.behindthecraft.com/ Where to find me: Newsletter: https://creatoreconomy.so/ X: https://x.com/petergyang LinkedIn: https://www.linkedin.com/in/petergyang/ Subscribe to this channel - more tutorials coming soon!

Claude Fable 5 is finally back, but there’s a catch: it’s only available on Claude subscriptions through July 12, and access stops once you hit 50% of your weekly usage limit. In this video, I walk through the five use cases I think are most worth spending Fable on: finding high-value work, getting life and business advice, making projects ship-ready, planning big features, and refactoring large codebases or skill systems. TIMESTAMPS (00:00) Claude Fable 5 is back, with limits (01:14) How to choose Fable 5 in Claude Code (01:30) Use case 1: Find Fable-worthy work (04:13) Use case 2: Get life and business advice (09:16) Use case 3: Make your project ship-ready (12:10) Use case 4: Plan the next big thing (15:54) Use case 5: Refactor a large codebase (18:29) Three tips to get the most out of Fable (19:57) Recap: the five best Fable use cases 📌 Get my personal AI OS with useful AI skills, $270 in AI tool credits, live monthly workshops, and courses on Hermes and Codex agents: https://www.behindthecraft.com/ Resources mentioned Claude: https://claude.ai/ Lavish editor: https://github.com/kunchenguid/lavish-axi Subscribe to this channel for practical AI tutorials and interviews.

The US still has the best frontier AI models. But China is competing across the whole stack: open-source models, energy, data centers, and chips. Restricting access is not enough. To compete, America needs the best closed and open-source models, plus much faster infrastructure buildout. 📌 Visit my YouTube for more videos.

How companies should think about driving AI adoption from Jess, Anthropic product lead. A lot of enterprises jump straight to automating giant cross-team workflows. Jess argues the better starting point is making every individual feel exponentially more powerful.

Here's how Anthropic PMs use agents internally to get closer to the product from Jess, product lead for Claude Managed Agents: “Access to our codebase has been the biggest unlock for me. It helps me manage state more easily. Rather than poking a bunch of engineers on what they’re doing, I can just track the PRs directly and see which ones are merged, which ones are deployed. I deeply understand and interact with my product so much more than I’ve ever been able to in the past.” 📌 Watch the full episode on my YouTube (link in profile)

As product lead at Anthropic, Jess has a first-hand look at how Anthropic teams are moving from prompting AI to building long-running agents that work overnight. In our episode, Jess showed me how to build a Claude analytics agent from scratch, then covered how Anthropic PMs use agents internally to understand the codebase, monitor customer feedback, and make better product decisions. Jess and I talked about: (00:00) AI agents that work while you sleep (03:46) Claude Managed Agents help for long-running tasks (05:22) Demo: Building a Claude analyst agent from scratch (09:28) Traces: How to see where an agent got stuck (15:11) Evals: How agents grade their own work (20:20) 5 ways in which Anthropic PMs use agents internally (28:06) Processing a 4,000-company waitlist with agents (32:50) To build your first agent, focus on helping one person (36:54) What changes when agents can work overnight Thanks to our sponsors: Linear: The AI agent platform for modern teams https://linear.app/behind-the-craft Wispr Flow: 4x faster than typing with your voice https://ref.wisprflow.ai/peteryang Riverside: All-in-one AI studio for podcasts and video https://creators.riverside.com/PeterYang 📌 Get the takeaways: https://creatoreconomy.so/p/inside-anthropics-bet-on-claude-agents-jess-yan 📌 Get my personal AI operating system with all my skills and prompts: https://www.behindthecraft.com/ Where to find Jess: X: https://x.com/jess__yan Website: https://platform.claude.com/docs/en/managed-agents/overview Subscribe to this channel - more interviews coming soon!

This is my complete course on how to set up the best AI chief of staff in the market today. It covers everything from install to integrations (e.g., email, calendar, Google Workspace) to cron jobs (e.g., morning briefing, health check). What I cover: (00:00) Why Hermes is the best 24/7 AI chief of staff (01:16) Where and how to install Hermes safely (02:55) The best AI model to pick for Hermes (05:04) Connect to Telegram for mobile chats (11:08) How Hermes uses tools, memory, and skills (12:39) How to personalize Hermes by editing its user and soul files (15:49) Two best ways to troubleshoot Hermes (18:15) Set up Hermes to respond via voice (22:47) Connect Gmail, Calendar, Docs, Drive, and Sheets step by step (38:25) Create routines like morning briefing, health check, and more 📌 Get the written tutorial: https://creatoreconomy.so/p/full-hermes-course-set-up-your-24-7-ai-chief-of-staff 📌 Get my complete Hermes course with all 17 lessons: https://www.behindthecraft.com/ Where to find me: Newsletter: https://creatoreconomy.so/ X: https://x.com/petergyang LinkedIn: https://www.linkedin.com/in/petergyang/ Subscribe to this channel — more tutorials coming soon!

From Bin: "A lot of the engineers here can make their own launch videos now because they understand the process, they understand the product. They work with their agent, and the agent makes everything after a lot of iterations. They don't have to learn a tool. They don't have to learn After Effects or even CapCut. Which is a huge unlocking for builders and entrepreneurs trying to build videos for their businesses but finding it extremely hard to learn." Full interview on @PeterYangYT!

From Bin: "We learned the hard way that agents, when working with JSONs and XMLs, have no visual intelligence. When an agent writes a JSON blob, it can be accurate, verifiable, and correct structurally. But the agent has no idea whether this JSON is going to be good looking or not. That is actually the biggest problem we ran into. That's when we turned to code. Code, especially HTML, is LLM's native language." Full interview on @PeterYangYT!

Bin and Jake built HyperFrames, HeyGen’s free open-source framework that lets AI agents create videos with HTML, CSS, and JavaScript. In this episode, they show how to turn any website into a launch video, then share the playbook HeyGen uses to make polished product videos fast. Bin, Jake, and I talked about: (00:00) From HTML to professional launch videos with AI (02:16) Getting started: Install HyperFrames for free (03:50) /website-to-video: Turn any URL into a launch video (08:28) Jake’s 5-step playbook to generate better videos (10:03) Step 1: Gather assets and create a frame.md (11:46) Step 2: Create a storyboard.md with the scenes (12:45) Step 3: Pull animations from HyperFrames launches (15:43) Step 4: Generate and review static frames (17:20) Step 5: Build the video and polish in Studio Thanks to our sponsors: Linear: The AI agent platform for modern teams https://linear.app/behind-the-craft Wispr Flow: 4x faster than typing with your voice https://ref.wisprflow.ai/peter-yang-jun21-pod Riverside: All-in-one AI studio for podcasts and video https://creators.riverside.com/PeterYang 📌 Get the takeaways: https://creatoreconomy.so/p/full-tutorial-make-professional-launch-videos-for-free-hyperframes-ai 📌 Get my personal AI operating system with all my skills and prompts: https://www.behindthecraft.com/ Where to find Bin and Jake: Bin: https://x.com/liu8in Jake: https://x.com/JakeFromHeyGen HyperFrames: https://github.com/heygen-com/hyperframes Subscribe to this channel - more interviews coming soon!

I’ve tried hundreds of AI apps. These are the 3 I keep coming back to: → Codex as the front door to my computer → Claude Code for front-end/design work → Hermes as my AI chief of staff If you have these three, you can get so much work done and save so much time. Check out my YouTube channel in my description for more deep-dive guides.

From Matt: "I have two projects that are a hundred percent finished and ready, and I'm afraid to ship them. It honestly makes me sad. I'm scared they won't be a hit, that no one will care and they'll only get a handful of retweets. The fear of rejection is real. But my advice is to just build and launch something anyway. It's okay. Even if you build something only for yourself, it's worth it. Even if no one ever used Agent Cookie, my own tool, I'd still get value out of it because it makes my agents so much better. That part is magical." Full interview on @PeterYangYT!

Today, I want to share 5 steps to build self-improving Claude Skills complete with evals and memory. I walk through building an /edit-post skill from scratch to edit any long-form writing. What I cover: (00:00) Why I built a personal AI advisor (02:09) Skill.md: How to define the advisor’s behavior (03:31) plan.md: The context file with goals, principles, energy, and life details (07:54) learnings.md: How the advisor remembers patterns over time (09:22) eval.md: What AI checks before giving me advice (10:15) Demo: Using Anthropic's Fable for deeper advice (11:31) How to build your own advisor skill now 📌 Get the written tutorial: https://creatoreconomy.so/p/how-to-build-an-ai-advisor-to-achieve-your-life-career-goals 📌 Get my personal AI OS with popular AI tool discounts and my best skills and prompts: https://www.behindthecraft.com/ Where to find me: Newsletter: https://creatoreconomy.so/ X: https://x.com/petergyang LinkedIn: https://www.linkedin.com/in/petergyang/ Subscribe to this channel — more tutorials coming soon!

I’m starting to share more of my AI builder journey here: → Interviews with my favorite AI builders → Practical AI tutorials for busy people → Honest thoughts on AI news → All the mistakes I make along the way Follow along if you’re building with AI too.

Matt has no CS degree, doesn’t know how to code, and doesn’t even read the plans that AI writes. Despite this, he’s built an incredible open source track record with projects like last30days (#1 trending on GitHub with 40K+ stars) and Printing Press (a tool that lets agents use any website or app). Matt has been a huge inspiration for my AI builder journey, and I think you’ll feel the same after watching our conversation. Matt and I talked about: (00:00) "I don't read code and the plans are for AI" (01:19) Compound Engineering: Matt’s go-to agentic loop (07:07) How Matt builds without a CS degree or reading code (11:18) Printing Press: Turning any website into an agent CLI (15:38) How Printing Press finds hidden APIs and GitHub projects (22:44) Why CLI + skill beats MCP for agent workflows (32:49) How Matt contributes to Python, Go, and other top repos (44:28) Why Matt is still afraid to launch finished projects Thanks to our sponsors: Linear: The AI agent platform for modern teams https://linear.app/behind-the-craft Wispr Flow: 4x faster than typing with your voice https://ref.wisprflow.ai/peteryang Riverside: All-in-one AI studio for podcasts and video. Use code PETERYANG to get your first month of Riverside Pro free https://creators.riverside.com/PeterYang 📌 Get the takeaways: https://creatoreconomy.so/p/the-ai-agent-system-behind-this-non-technical-founder-matt-van-horn 📌 Get my personal AI operating system with all my skills and prompts: https://www.behindthecraft.com/ Where to find Matt: X: https://x.com/mvanhorn GitHub: https://github.com/mvanhorn Printing Press: https://printingpress.dev last30days: https://github.com/mvanhorn/last30days-skill Subscribe to this channel - more interviews coming soon!

Give yourself permission to build. The traditional career ladder pushes everyone to become a leader, but I want to be a builder. I know a lot of builders who spent their best years climbing the wrong ladder. If you're a builder at heart, embrace it.

Last year, I made seven figures at my product job. Last week, I left that job to bet on myself. This wasn’t an easy choice to make. After over a decade in product, walking away felt like giving up a core part of my identity. Here are 5 principles that helped me make the decision after alot of reflection. What I cover: (0:00) Why I quit my seven-figure product job (0:23) Go where work feels like play (3:44) Give yourself permission to build (5:35) At some point, trading time for wealth just isn't worth it (8:33) Surround yourself with people who share your curiosity (10:11) The people you love won't be there forever (13:57) What I'm doing next (and a promise) 📌 Get the written post: https://creatoreconomy.so/p/i-quit-my-high-paying-product-job-to-bet-on-myself 📌 Get my personal AI OS with popular AI tool discounts and my best skills and prompts: https://www.behindthecraft.com/ Where to find me: Newsletter: https://creatoreconomy.so/ X: https://x.com/petergyang LinkedIn: https://www.linkedin.com/in/petergyang/ Subscribe to this channel — more tutorials coming soon!

This is something I think is going to fundamentally change as we progress on AI adoption. Our workflows and how our teams work were built at a time when we spent most of our time coding. The average software engineer writes 10 to 15 PRs a month. When that's the case, it's fine for everyone else to do code reviews and run all these processes. But when you start writing 10 times more PRs, we're not ready for that. Our processes, our team composition, everything was built without that assumption in mind. So things are starting to break. A lot of teams are changing their practices to fight that. That's the kind of change I'm starting to see. Full interview on @PeterYangYT!

From Kun: "AI can write so much code that if you review every single line, you become the bottleneck. So I don't even review the first-pass code from the agent. I use a tool I built called No Mistakes. I did a lot of prompt engineering to get the agents to scrutinize the change very hard, so any edge case, bug, or logical error gets caught. It's a very high recall phase. When I first built it, I ran parallel testing. I let the agents review the change, and I reviewed it myself, to see how often I caught something they didn't. I used that to iterate on the prompts and the workflow. Eventually, I got to a point where I find myself never catching anything the agents don't catch." Full interview on @PeterYangYT!