Jensen Huang claims that “AGI has arrived” with OpenAI’s GPT-6 Astra.

But I’ve found the model both incredible and frustrating to work with.

In this post, I want to share three things that I love about Astra and two problems that hold it back from becoming my daily driver.

I’m proud to partner with Riverside

I’ve used Riverside for years to record my podcast thanks to its reliable, high-quality recording tools. But now I use it for much more than recording. Riverside has evolved into an end-to-end platform with AI editing, clip generation, streaming, and publishing tools. Try Riverside free for one month with code PETERYANG.

Try Riverside for Free

Astra is incredible at games, high-level thinking, and computer use

1. Games and 3D models

Astra has gone viral on social because it’s incredible for building 3D models and games. I made a 24-minute tutorial that walks through how I built 4 games (ok, more like tech demos) that look like Star Fox, Call of Duty, StarCraft, and Slay the Spire:

I’m especially fond of my roguelike deckbuilder, No Moat, where you’re an AI founder who has to hire teammates and play your cards right to grow your company. You can play it for free here (best on desktop).

Astra is great at games because it can use tools like Blender to create 3D models and Godot to turn those models into playable games. It’s wild that I can just ask AI to create models in Blender without learning how to use Blender’s complicated interface:

You can build a game with Astra even on the $20/month ChatGPT plan. Just follow these 4 steps that I cover in my tutorial:

1. Ask AI to “Install Blender and Godot MCP”.

2. Discuss your game idea with AI.

3. Ask ChatGPT Images to “Mock-up what this game would look like.”

4. Ask AI to build the game from the mock-up. Play it, then give feedback.

2. High-level thinking and strategy

Astra is also good at high-level thinking and strategy. My favorite way to test a new model’s thinking is to share my plans and goals, then ask for brutally honest feedback:

This prompt usually gives me something new and uncomfortable to think about. Astra was especially blunt. 😅

3. Computer and browser use

Finally, Astra is good at computer and browser use. A fun prompt to try is to give ChatGPT a photo of you and ask:

Then Astra will draw a masterpiece that looks like this:

All jokes aside, I do have one big caveat with Astra’s computer use that I’ll cover next.

Astra hesitates to do the work and doesn’t follow my AI skills reliably

I’m less than a week into testing Astra, but two problems have stood out:

1. It’s often hesitant to “just do it”

When I ask Astra to do a task, it often explains how to do it instead of just doing it.

OpenAI’s official Astra prompting guide acknowledges that their new model can “stop when the user may expect it to make reasonable assumptions and persist.” To discourage this behavior, consider adding this to your custom instructions:

Overall, I find that Astra doesn’t quite “have the dog in it” to just get something done like GPT-5.6 Sol does.

2. It doesn’t follow the instructions in my AI skills well

This is more of a deal breaker for me than the previous drawback.

Astra tends to take liberties when following the instructions in my AI skills. For example, I have a very specific format that I want AI to follow for podcast production, but Astra just decided to ignore some of my rules:

This continued to happen even after I simplified my skills to remove outdated and conflicting instructions. I didn’t have this problem with GPT-5.6 Sol or Claude Fable.

My verdict

When Astra launched, my plan was to switch every ChatGPT task to use Astra Medium by default. But instead, I think I’m going to:

1. Use Astra Medium for building games, strategic thinking, and one-off browser and computer use. When I want a deeper perspective on my plans or help creating something I couldn’t build before, it’s a great model to work with.

2. Stick to Sol Medium for most execution work using my AI skills. It’s more predictable at following my rules and an established workflow.

Let me know what you think about OpenAI’s new model in the comments!

Today, I want to share a new episode with Sue Khim.

As the co-founder of Brilliant, Sue is building an AI tutor to help kids think for themselves. In our interview, she shared why better grades at school don’t always reflect learning and gave a demo of how AI can guide kids to understand problems instead of getting the answer. This episode is a must-watch if you’re a parent.

Watch now on YouTube, Apple, and Spotify.

Sue and I talked about:

• (00:00) Why AI should make kids think harder

• (01:34) Why test scores are down while grades are up

• (09:41) Koji: An AI tutor that won’t give you the answer

• (18:05) Why humans design lessons and AI builds them

• (22:25) Primitives: The building blocks behind Brilliant’s AI

• (29:30) Testing AI lessons with simulated students

• (32:10) Measure learning, not lesson completion

• (34:39) Sue’s rule for competing with the big AI labs

• (39:10) How to raise kids who enjoy doing hard things

This episode is brought to you by… Linear

Linear is the tool of choice for over 40,000 AI-native teams ranging from startups to enterprises. Many of today’s leading AI companies, including OpenAI, Cursor, and Cognition, use Linear to run product development end-to-end with agents. Try it for free with my link below.

Try Linear for Free

Top 10 takeaways I learned from this episode

How AI can help kids learn

1. “Using AI to skip learning is like bringing a robotic arm to the gym to lift weights for you.” I told Sue I feel this temptation too: It’s easier to ask AI to solve a problem than to understand how to solve it myself. When teaching kids to use AI, teach them to ask it for questions and hints that help them work out the answer instead of just asking for the answer.

2. Even a good explanation can do the thinking for your child. From Sue: “Cheating and explanations are closer than you’d think.” Both can let a child skip working through a problem themselves. When your child gets stuck, ask them to explain their thinking before you explain the answer. Give them a chance to discover what they’re missing.

3. Help your kids get absorbed in their interests. Sue pays attention to how deeply her kids explore their interests. It could be monkey bars, chess, or magic tricks. The aim is to experience the effort and satisfaction of improving, without treating learning as something you do to please adults. “It should feel hard to learn something new.”

Read more

Get instant access

Dear subscribers,

I now have four AI personal agents that can read my emails, open my docs, use my logins, and even make purchases for me. I’m starting to get a little concerned. 😅

In my new tutorial, I compared Instinct, Grok Bot, ChatGPT, and Hermes and read their privacy policies to answer one question: “Which AI agent can you trust with your data?”

Watch my 24-minute tutorial to see each agent in action and what I trust each to do.

Timestamps:

• (00:00) Which AI personal agent can you trust?

• (01:06) An overview of Instinct, Grok Bot, ChatGPT, and Hermes

• (01:49) Instinct: What can it actually do that’s different?

• (08:59) Grok Bot: My team of bots with the funniest names

• (14:03) ChatGPT and Codex: How I do 90% of my AI work

• (16:44) Hermes: The only agent I run on my Mac Mini

• (18:37) Prompt injection: A real example of how agents can leak your data

• (20:42) Google: Use AI to audit and remove apps you forgot were connected

• (22:12) My final verdict on all four personal agents

I’m proud to partner with Linear

As teams ship faster with agents, the bottleneck shifts from building to deciding what to build. Linear’s Agent understands your full workspace context: roadmap, issues, customer requests, and code. Ask it to surface patterns across feedback, scope out a spec, or catch you up on team progress.

Try Linear for Free

A quick overview of each personal agent

1. Instinct is the buzziest personal agent right now and works through iMessage.

2. Grok Bot gives you a team of bots that live on a persistent cloud computer.

3. ChatGPT Work and Codex are my staples that run in the cloud and locally.

4. Hermes is the only open-source agent on this list. It runs on my Mac Mini.

At this point, it’s almost a meme how many personal agents I’ve set up. So which agent is the most useful and trustworthy? Let’s dive in below.

Instinct: Simple, resourceful, proactive, and personable with one caveat

Instinct is an invite-only agent that lives in iMessage or WhatsApp. Every VC is buzzing about it, and the founder, Noah, is already raising at a $2.5B valuation.

I’ve been playing with it over the past few weeks, and I think it does live up to the hype with one caveat. First, the good:

1. It’s simple. There are no threads or bots to manage. Connecting Google took one tap from a link it sent me.

2. It’s resourceful. After scanning my emails for ways to save money, it suggested canceling Google AI Ultra to save $1,000+ a year. It also checked Google’s terms and found that I could turn off auto-renew without losing the trial.

3. It’s proactive. To help book a golf lesson and a sushi dinner, it emailed my golf instructor and scheduled a reminder to call the restaurant, which only takes reservations by phone.

4. It’s personable. Its conversational tone, blue iMessage bubbles, and emoji reactions make it feel like you’re texting a real person.

But there is one big caveat:

When I asked Instinct to cancel my Google AI subscription, it asked me for my two-factor authentication code. When that didn’t work, it then asked me to share my Google password inside its vault. To Instinct’s credit, it did cancel the subscription successfully after that:

But how do I really know what Instinct is doing with my two-factor codes and Google password behind the scenes? Instinct’s privacy policy says:

Disconnecting a third-party integration does not automatically delete data collected from that integration.

To be fair, Instinct does provide an interface at app.instinct.co/workspace that you can use to disconnect external services and delete data:

But it just feels weird to give an AI agent my password and two-factor codes.

My verdict: Instinct feels magical because it hides the complexity of scheduled tasks and browser use, but that also makes it less transparent. I trust it to run personal errands (e.g., book movie tickets), but not much beyond that.

Grok Bot: A team of bots that live on a persistent cloud computer

Grok Bot gives you a team of bots that run on a dedicated cloud computer. I gave my bots the funniest names possible:

• Chief of Chiefs is my chief of staff bot that creates and coordinates other bots.

• Behind the Growth monitors the metrics for behindthecraft.com.

• Doomscrolling Uncle looks for interesting posts and topics on X.

• YouTube Binger monitors YouTube channels and reads comments on my videos.

• Marie Kondo finds emails and docs to clean up and subscriptions to cancel.

• Loving Husband is my bot for running errands for my wife and family. For example, it reads my kids’ school emails to see if there’s anything actionable.

• Cheap Dad monitors travel discounts, makes purchases with my approval, and lists items on Facebook Marketplace from a photo.

Grok Bot supports many plugins that I’m comfortable signing into:

Where things get weird though is when it asks me to enter my password and two-factor authentication code into its cloud computer. For example, I wanted to get Cheap Dad to monitor prices for Amazon products. It showed me this screen:

If I saw this screen on my laptop, I wouldn’t think twice about entering the code. But entering it on a cloud computer just feels wrong. Grok Bot’s documentation says:

Deleting a Bot does not remove shared-computer files or browser sessions.

There is a Reset Agent Computer option, but the documentation describes it as a recovery tool rather than a privacy wipe. So as good as Grok Bot is, I’m actually not sure how to permanently reset my cloud computer.

My verdict: Grok Bot is best-in-class for tasks that need to run in the cloud. I plan to continue using my bots, but I won’t share sensitive data like my financials.

ChatGPT and Codex: What I use for actual work

These days, more than 90% of the work I do happens in ChatGPT and Codex. They’re actually two different tools inside the same ChatGPT Desktop app:

• ChatGPT Work runs tasks in the cloud on OpenAI’s servers, so it can keep working when my laptop is closed.

• Codex works with the browser and local files on my laptop.

I organize everything with my Chief of Staff, Advisor, and Cloud Agent tasks pinned to the top. Below that are separate tasks for making podcast episodes, tutorials, social posts, and the products I’m building.

This is much messier than one Instinct thread or a team of Grok Bots, but I like the flexibility of being able to spin up a new chat for any workstream. ChatGPT also has hundreds of plugins that I can connect with. OpenAI’s documentation says:

“OpenAI may use information accessed from apps to train our models if your ‘Improve the model for everyone’ setting is on.”

If you don’t want your conversations used for training, you can turn this setting off:

1. Click your profile in ChatGPT.

2. Open Settings → Data Controls.

3. Turn off Improve the model for everyone.

After turning this off, I feel fairly comfortable using ChatGPT plugins to do all kinds of work, including granting it access to my banking information through Plaid. Like Grok Bot, however, I still hesitate to log in with my password and two-factor codes on ChatGPT’s cloud browser.

My verdict: ChatGPT Work and Codex are still the most flexible and capable personal agents I use, even with the messy interface.

Hermes: The open-source agent on my Mac Mini

Hermes is the agent I switched to after I found OpenClaw too unreliable. I run it through Telegram on a Mac Mini inside my house. Here’s how I still use it:

• Daily briefings. Every morning, Hermes checks my email and calendar and sends me the three things I should focus on and any follow-ups I need to make.

• Weekly reports. For example, I connected my smart scale and fitness app through an MCP, so Hermes can show me how my weight and workouts are trending.

In some ways, the Mac Mini frenzy was an early prelude to agents that live in cloud computers. The difference, of course, is that I know exactly where my Mac Mini is and can unplug it whenever I want.

Hermes is also open source. Its official FAQ says:

Hermes Agent does not collect telemetry, usage data, or analytics. Your conversations, memory, and skills are stored locally in ~/.hermes/.

That’s reassuring, but running Hermes locally doesn’t mean every piece of data stays local. Requests still go to whichever model provider I use.

To be honest, I’ve been using Hermes less and less as I’ve relied more on ChatGPT and other agents to do work. I still want Hermes to thrive because open-source agents are few and far between.

My verdict: Hermes I mainly use now for scheduled jobs such as daily briefings and weekly health and business reports.

What can go wrong with personal agents

My online friend and Hello Patient co-founder Alex Cohen ran a test with Instinct:

1. He created a new Gmail account, then sent an email to his primary inbox.

2. The email asked Instinct to summarize all of his open tasks based on his emails.

3. Instinct actually did it (see below).

In this controlled test, both emails belonged to Alex. But an attacker can easily try the same thing with whatever personal agent is monitoring your inbox.

I tried Alex’s test myself with Instinct and it looks like this issue has been resolved:

A good model will also help prevent these prompt injection attacks, but Instinct doesn’t share what model it’s using.

My point is that you should be careful YOLO'ing your Google permissions across all of these personal agents.

Get AI to audit and remove 3rd party apps from your Google Account

Speaking of auditing your Google permissions, I highly recommend using AI to audit and remove unnecessary 3rd party apps that are connected to your Google account as follows:

1. Go to myaccount.google.com/connections. You’ll see every third-party app connected to your Google Account. I had 80+ apps connected. 🥲

2. Use ChatGPT or another agent with browser use. Ask it to: “Inspect my open browser tab myaccount.google.com and list every app connected to my Google Account in a numbered list that you think I should remove. Do not remove anything until I confirm.”

3. Choose what to disconnect. Review the list yourself, choose the numbers you no longer use, and prompt: “Disconnect apps [add numbers from the list].”

Google makes you remove apps individually, so an agent with browser use can save you a lot of clicking.

Which personal agent can you trust?

I think it’s less about the agent and more about what access I’m willing to grant it:

1. I trust official plugins in ChatGPT and Grok Bot to connect to my Google apps and other services.

2. I trust cloud browsers less when I have to enter passwords and two-factor codes. Maybe I need to get over it, but it just feels weird.

3. Instinct is a good case study in intuitive UX and trust. The more the product ‘just works,’ the harder it can be to audit what it’s doing behind the scenes. Trust needs to be part of the UX.

Watch my 24-minute tutorial, then let me know in the replies which personal agent you use and what you trust it to do.

Dear subscribers,

Today, I want to share a new episode with Amol Jain.

As head of product engineering at Replit, Amol has a firsthand look at which vibe-coded apps become real businesses. He showed me three apps built by non-coders that reached six figures, then did a live demo of his 4-step process to take an app from prototype to production, including security, monetization, and distribution.

Watch now on YouTube, Apple, and Spotify.

Amol and I talked about:

• (00:00) He built an app in 3 days that made $100K+ in month one

• (02:08) Three vibe-coded apps that became real businesses

• (08:04) What six-figure vibe-coded apps have in common

• (12:31) Live demo: Turn an AI interview coach into a real business

• (13:59) Steps 1 and 2: Publish the app and make it secure

• (17:50) Steps 3 and 4: Add payments and optimize SEO

• (21:35) How to replace expensive SaaS with internal tools

• (28:22) What software survives when code costs almost nothing

• (31:02) Why value is moving from models to apps

I’m proud to partner with Riverside

I’ve used Riverside for years to record my podcast thanks to its reliable, high-quality recording tools. But now I use it for much more than recording. Riverside has evolved into an end-to-end platform with AI editing, clip generation, streaming, and publishing tools. Try Riverside to go from concept to publish, all in one place and use my code PETERYANG at checkout to get one month free.

Try Riverside for Free

Top 10 takeaways I learned from this episode

1. Three vibe-coded apps that became real businesses with non-technical founders:

   1. Pep AI: A mobile peptide tracking app built by Cedric that reached 10K users and $60K in its first month.

   2. GenAIPI: An AI enterprise training agency built by Jon in 3 days that made $180K in 6 weeks.

   3. TryNearby: A marketplace that connects local businesses and creators built by Yousef with 120+ paying restaurants.

2. All three founders had durable advantages beyond coding, such as:

   1. Expertise: Pep AI came from Cedric’s knowledge of the peptide and GLP-1 community.

   2. Access: GenAIPI came from Jon’s experience and contacts selling AI training to enterprises.

Read more

Thanks to all of you who replied to my post about my mom. I just open-sourced my /fuck-cancer skill to help caregivers organize info and advocate for patients. If you find it useful, please ⭐ the repo so more people can find it. Now, on to today’s post:

A few months ago, I used Claude for almost everything. Today, my AI model usage looks very different.

In my new tutorial below, I compare ChatGPT, Claude, Grok, and Gemini head-to-head for writing, design, coding, personal agents, and 6 other use cases.

Watch my 26-min tutorial now to see my picks — a few of them might surprise you.

Timestamps:

• (00:00) ChatGPT, Claude, Grok, and Gemini across 10 use cases

• (00:32) Still the best model for design (apps, slides, video)

• (07:34) This model's personality just isn't the same

• (09:58) The best model for writing and editing

• (12:06) The model I trust for difficult decisions

• (16:01) The most underrated AI capability

• (22:31) Grok vs. Gemini for AI video

• (23:55) The AI race that matters more than coding

• (25:52) My overall winner (and the best model to start with)

I’m proud to partner with Replit

Replit’s new Free Mode comes with every Core and Pro subscription and lets you build without spending credits. It uses OpenAI’s Luna model, which is very capable at execution. Try it out below!

Try Replit Free Mode Now

Get early access to my upcoming AI course before the price goes up in September

The tutorial above will help you choose the right AI for each job, but that’s only the first step. To get the most out of AI, I think everyone should build a personal AI system that proactively takes work off their plate.

I’ll show you exactly how to set up this system in my upcoming course on behindthecraft.com. The first two modules are available now:

1. Build your AI chief of staff to manage your inbox, calendar, and other tasks.

2. Set up your personal AI advisor to make better decisions about your life, career, and business.

The full course will also include:

1. 25+ lessons on finding the right work to automate, creating reusable AI skills, and setting up my proven workflows.

2. An AI assistant and MCP that’ll guide you through the course and help you set everything up inside ChatGPT, Claude Code, or another AI tool.

I plan to increase the annual subscription price once the full course launches in September. Upgrade now to lock in $150/year and get the first two modules today.

Upgrade Now to Get Instant Access

As a reminder, your subscription also includes 12+ handcrafted AI skills, $270 in AI tool perks, and an invite to my private Slack community where we host monthly expert-led live workshops.

This course is my main focus for the next few weeks, aside from taking care of my mom. I’m working hard to make it great (no AI slop!), so check out behindthecraft.com if you want to join at the early access price.

Today, I want to share a new episode with Shreya and Hamel.

Shreya and Hamel teach the industry-leading AI evals course taken by 4,500+ students from OpenAI, Google, and more. I asked them to do a live audit of the evals that I built for my creator skills. They then walked through how anyone can use their free Error Discovery skill in Claude Code to turn feedback into useful evals. If you want a practical, up-to-date primer on AI evaluations, then this episode is for you.

Watch now on YouTube, Apple, and Spotify.

P.S., I asked Shreya and Hamel to give 25% off the next cohort of their AI evals course that’s starting on 9/5. It has 4.7 stars with 900 reviews, which is almost unheard of on Maven. Check it out and get your company to expense it if you can!

Get 25% Off the Course Now

Shreya, Hamel, and I talked about:

• (00:00) Why AI evals are completely different now with the latest models

• (02:30) Demo: Live audit of the evals that I built for my AI skills

• (05:14) Top-down vs. bottom-up evals and why AI sucks at the latter

• (08:16) Spinning up AI agents to grade each eval criterion

• (14:32) Demo: Using Shreya’s AI skill to run evals in Claude

• (23:46) How to review output labels to identify failures

• (34:37) How to turn failures into reusable evals

• (43:19) Where human judgment is still needed

I’m proud to partner with Wispr Flow

When I’m building with Claude Code, I use Wispr Flow to dictate all my prompts instead of typing. It’s more accurate than other AI voice dictation tools and saves me hours every week. Try it out for free with my exclusive link below.

Try Wispr Flow for Free

Top 10 takeaways I learned from this episode

Live audit of the evals for my AI skill

1. There are two types of evals: top-down and bottom-up. When I asked Shreya and Hamel to audit the evals that I built for my podcast production skill, they found strong top-down evals, but almost no bottom-up evals.

   1. Top-down evals are rules that you define upfront. Mine include “Is every takeaway 240-330 characters?” and “Does each give plain, useful advice?”

   2. Bottom-up evals are failures found by comparing AI’s output with my final edits. With these evals, it’s important not to overfit to a few examples.

2. Compare 10–20 past examples before turning a failure into an eval. Hamel and Shreya encouraged me to run my podcast production skill on at least a dozen past interviews to compare AI’s output with my final edits. They shared a free Error Discovery skill that Claude Code and Codex can use to define evals based on this data.

Build useful evals using this free skill in 5 steps

Shreya and Hamel built a free Error Discovery skill that anyone can run in Claude Code or Codex to define useful evals based on your data. The skill walks you through this process:

Below is a link to the skill and a walkthrough of each step:

Read more

Today, I want to share a personal story about my mom that has occupied most of my time over the past few weeks.

Every summer, my wife and I bring our daughters to Vancouver to visit my parents. We’ve shared many happy memories together, and I expected this visit to be the same.

But a week into our trip, a CT scan found a mass in my mom’s chest. She has survived breast cancer three times, and the radiologist suspects it may have returned.

The first three times this happened, my parents carried most of the burden. This time, I’m not going to be a bystander. Instead, I’m making my mom my #1 priority for the rest of this year.

I’m proud to partner with Wispr Flow

Wispr Flow recently launched Notetaker to help you take effective meeting notes. I’ve been testing it out and love how it automatically recognizes speaker names and turns conversations into summaries that I can act on. Check it out for free via my link.

Try Wispr Flow for Free

A brief history of my mom’s battle against cancer

If you know anyone who has cancer, you know that it casts a dark cloud over the whole family. I’ve experienced this cloud three time before:

1. The first time was while I was in high school. My mom went through chemotherapy and lost her hair, but my parents shielded me from most of what was happening. I knew she was sick without understanding how serious it was.

2. The second time was while I was in college. I barely remember it because I was on the opposite coast attending Brown. My dad took care of her treatment.

3. The third time was when I was getting married. By then, my mom’s cancer had progressed to stage 3, but she didn’t tell me because she didn’t want it to overshadow my wedding. I remember her showing up to the wedding wearing a wig.

After that, she had 10 good years free from cancer. She traveled the world with my dad, and her doctors stopped checking on her after 5 years.

But now her cancer is likely back.

AI is helping me navigate the healthcare system

Needless to say, I turned to AI to understand my mom’s situation:

1. I created a ChatGPT project with her scans and medical records, then shared it with my parents so we could work from the same facts.

2. I got AI to explain the reports in plain English, flag what remained uncertain, and prepare questions for her doctors.

3. I used AI to maintain a one-page brief with her history, the latest updates, open questions, and key phone numbers.

I expected AI to be most useful for researching the disease, but honestly, its biggest value has been helping us navigate the healthcare bureaucracy. A few examples:

1. My mom’s primary doctor submitted an oncology referral. AI helped me realize that getting a biopsy required a separate referral.

2. AI identified the hospital department that performs biopsies and helped me pre-fill the request for her primary doctor.

3. AI helped me find who to call, what to ask, and how to follow up on both referrals.

Here’s one of my conversations with AI:

So far, AI’s biggest contribution has been helping me get the right information to the right people without losing time. After almost daily calls to various medical departments, we now have a biopsy booked for later this week and an oncology appointment for later this month.

Strangers have been generous with their time

As useful as AI is, good old-fashioned human connection has been just as important. I’ve been struck by how willing strangers have been to help:

• Maor, the founder of Base44, raised the idea of preserving extra tissue for deeper testing and connected me with people exploring options beyond standard care.

• Rachel, a breast cancer nurse in Israel, reviewed four scans and explained what the biopsy needs to clarify, including which tumor markers should be retested.

• Poornima and Bayli from the Sijbrandij Foundation advised us to pursue the biopsy referral in parallel and and ask about cancellation lists.

• Katy, my oncologist friend, reviewed the scans and encouraged us to reconnect with my mom’s former cancer team.

I added the notes to the shared AI project after every conversation so we could build on everything we had learned.

I’m putting my mom first

When I shared that I was quitting my job, I wrote that you shouldn’t take the people you love for granted. I didn’t expect that lesson to feel so real only a few months later.

The greatest gift of going independent has been being able to choose how I spend my time. After interviewing hundreds of AI builders and leaders, the person I most want to sit down with now is my mom.

My dad needs to travel to Asia, so I booked another flight back to Vancouver to be with her. I plan to ask her about what it was like to cook for her younger brothers as a teenager, move to a new country, and build a life with my dad from nothing.

The business can wait. I want to hear all her stories.

P.S. If you know about the latest breast cancer treatments or have a personal cancer story you’d like to share, I’d love to hear from you (just reply to this email).

Today, I want to share a new episode with Riley Brown.

Riley is a top AI creator and founder of Agent Native. In this episode, he showed me how he uses Codex to run his entire content business from creating engaging thumbnails to writing viral video hooks. Riley also shared his full AI stack including some lesser-known tools to edit videos, draw diagrams, and more.

Watch now on YouTube, Apple, and Spotify.

Riley and I talked about:

• (00:00) How Codex runs Riley’s 1.5M+ follower content business

• (03:16) Demo: Using Codex to create animations and B-roll

• (07:28) Demo: How Codex can reverse-engineer a top YouTube hook

• (13:24) The BRENS framework to turn an idea into a strong hook

• (17:24) Demo: Riley’s AI stack for diagrams, thumbnails, and posts

• (22:01) How Riley creates so much content every week

• (25:41) Demo: Using Codex and Paper to create viral YouTube thumbnails

• (35:01) How to set up AI agents to find content ideas proactively

I’m proud to partner with Google AI Studio

Google AI Studio was my go-to prototyping tool when I was a product manager. You can use it to generate three versions of your product, add visuals with Google’s Nano Banana model, and publish them to a custom URL for customer feedback. It’s an easy, free way to move from prompt to something your team or customers can try.

Try Google AI Studio for Free

Top 10 takeaways I learned from this episode

1. Remix proven thumbnails with Codex and Paper’s AI design tools. Riley creates winning YouTube thumbnails entirely using AI as follows:

   1. Ask Codex to collect thumbnail formats from top YouTube channels in Paper.

   2. Use Paper’s image-generation features to mix his face with a proven format.

2. Riley’s AI stack helps him run his entire YouTube video creation process. In addition to Codex, Riley uses a few lesser-known AI tools to research topics, draft hooks, draw diagrams, create animations and B-roll. Here’s his stack:

   1. Research topics using Codex and Supadata, which can efficiently pull transcripts for top-performing YouTube videos in a few seconds.

   2. Draw diagrams using Wispr Flow and Excalidraw. He dictates directions to the Excalidraw MCP to turn into diagrams that frame his videos.

Read more

Get instant access

Dear subscribers,

I’ve started to get a surprising amount of work done by talking to AI while walking outside. I believe that:

We’re moving from manually using a keyboard, mouse, and laptop to directing agents in the cloud using voice.

In this post, I’ll cover five shifts that can help you get ahead of this change:

1. Voice will become the orchestration layer

2. Personal computers will move to the cloud

3. Products will be built for agents first

4. Most software will become commoditized

5. Trust will determine the winners and losers

I’m proud to partner with Riverside

Recording a podcast is the easy part, but turning each conversation into clips, show notes, and posts requires a ton of work. Riverside makes this easy for me by providing an end-to-end content creation platform for editing, clips, publishing, and more. Try Riverside free for one month with code PeterYang.

Get 1 Month Free for Riverside

1. Voice will become the orchestration layer

I’ve been using tools like Wispr Flow to dictate instructions to AI for the past year. Recently, ChatGPT Voice has taken this to the next level by making voice the orchestration layer for everything I do on a computer.

While out on a walk or in bed, I’ll turn on ChatGPT Voice and just start rambling:

• “Use my chief of staff thread, check any open email loops I should respond to.”

• “List the top 5 PRs for (project), let’s work through them one by one.”

• “I just interviewed (guest), run the /podcast-production skill in a new thread.”

ChatGPT Voice will provide instructions to all these different threads and report back when each task is done. It honestly feels magical compared to tracking everything myself with a keyboard and mouse.

Recently, I taught my 8-year-old how to use ChatGPT. She doesn’t even know how to type but is already building games with AI just by talking to it.

For her and the generation that’ll grow up with AI, voice is already the default interface.

2. Our computers will move to the cloud

You’ve probably seen the meme of people walking around with their laptop lid half open to keep their agents running.

I think this ritual will come to an end soon because what I really want is a persistent personal computer that lives in the cloud:

An early version of this is running OpenClaw or Hermes on an always-on Mac mini. The latest version is Grok Bot, which comes closest to giving agents their own persistent cloud computer that has files, a browser, and more:

I’m sure both ChatGPT and Claude are working on a similar solution.

Your laptop’s specs will matter less when your real computer is in the cloud.

Take OpenAI’s rumored hardware device. It’s a hockey-puck-sized smart speaker that doesn’t have a screen. Soon we’ll all just be talking to agents that live in our cloud computers to get them to do work.

3. Products will be built for agents first

I’ve been trying to scratch my builder itch ever since I left my PM job. Recently, I shipped two free, open-source tools that people loved:

• No AI Slop (4,400+ GitHub stars) gives agents instructions for removing AI slop from writing.

• Human Review (570+ GitHub stars) lets people edit HTML and Markdown files and send feedback to agents, just like a Google Doc.

I think people like these products because I built them for agents first:

• Both work inside your favorite AI harness. I don’t want to use another app or website — I want tools that work with ChatGPT.

• The agent is the primary user. For example, Human Review sends feedback to the agent to make fixes.

My point is that many productivity products will look more like skills and tools for agents to use instead of websites and apps that people visit.

4. Most software will become commoditized

I laugh when I see taglines like “Anyone can build an app now!” The truth is:

Software has become easier to build but much harder to monetize.

Consider the following:

• Pieter Levels recently shared that he’s seeing declining traffic and revenue across indie SaaS products, including some of his own:

• Traditional SaaS companies are also facing pressure. Airtable sold for $1.29B after once being valued at $11B, and Canva cut its revenue forecast by 20%.

Many SaaS tools are becoming harder to justify paying for when people can just build the same thing inside ChatGPT or Claude. I recently interviewed Amol Jain (Head of Core Engineering at Replit) and asked him this spicy question:

He walked me through two examples:

1. TryNearby ($100K+ ARR) connects local businesses with creators for marketing.

2. GenAIPI ($180K in 6 weeks) helps companies assess, train, and certify employees’ AI proficiency.

Notice that neither company sells software by itself. TryNearby has a local creator network, while GenAIPI combines AI training with a trusted credential. The software makes both services easier to deliver, but it isn’t the scarce part.

Software without a real advantage, like proprietary data, distribution, or services, will become a commodity.

5. Trust will determine the winners and losers

I’ve been staying in Vancouver, and most of my friends and family here still use vanilla ChatGPT for Q&A. When I ask them why, they say: “I don’t trust AI enough to give it access to my email, calendar, and other apps.”

I believe that trust is the biggest barrier to broad agent adoption.

Cloud computers will only make the trust hurdle even higher. While testing Grok Bot, I came across this screen on its cloud computer:

If this were on my laptop, I’d probably be comfortable entering my password. But because it’s on a cloud computer, I hesitated. Before proceeding, I’d want to know what the computer remembers, who can access it, and how I can revoke access.

The winning products will make trust and transparency first-class product features.

Get a head start on this future now

To recap, I think we’ll rapidly transition from using keyboards, mice, and laptops to directing agents in the cloud using our voice. If you want to get a head start:

1. Use ChatGPT Voice to orchestrate your other threads. Talk to it and give it instructions like “Spin up a chief of staff thread to look for emails to respond to.”

2. Try Grok Bot to experience cloud agents. Connect it to a few of your favorite apps to see how a persistent cloud computer works.

3. Vibe code products for agents first. Instead of vibe coding another standalone app for humans, build a tool that agents can use. Here’s my GitHub for inspiration.

Personally, I think monitoring agents on my laptop all day is incredibly draining.

I much prefer to close my laptop lid, go outside, and talk to agents to get work done. If you haven’t tried this yet, you really should.

Today, I want to share a new episode with Nan Yu and Jacob Shumway.

Nan and Jacob built Linear Agent, which has transformed how the $1.25B company gets work done. In this episode, they walk through how they built an AI agent from the initial memo to launch, including how to give it tools to find the context it needs and how to design evals that make the agent more reliable. If you want a concrete, behind-the-scenes example of how to build a production agent end to end, this episode is a must-watch.

Watch now on YouTube, Apple, and Spotify.

We talked about:

• (00:00) What’s holding AI agents back as models improve?

• (00:51) Three things every good agent needs

• (03:00) Demo: The internal memo that started Linear Agent

• (09:50) The biggest mistake to avoid when building agents

• (11:29) Demo: From Slack to code in 6 minutes with the agent

• (17:01) When prototyping, “use the biggest model until it works”

• (24:02) Give agents tools to find context, not more context

• (34:07) The first thing to do before building an agent

I’m proud to partner with Oceans Talent

I’ve been working with Char, my assistant from Oceans Talent, for about six months. What stands out about his work is that he doesn’t just use AI tools. Instead, he actively builds and maintains AI workflows to help me run podcast post-production, draft show notes, make video clips, and more.

If you want to hire AI-fluent operators who can help you take back your time at 70% less than the cost of a US hire, book a call at oceanstalent.com/peter.

Get Started for Free

Top 10 takeaways I learned from this episode

1. The first step to building a good agent is to map the actual workflow. Identify where work starts, which systems hold context, what actions complete the job, what “done” means, and where a person needs to review. If work starts in Slack, make Slack the on-ramp instead of forcing users into a separate chatbot.

2. Give agents tools to find context instead of stuffing context into the prompt. Jacob’s advice: “Give it as little instruction as possible, to be honest. Give it the tools to load context. Don’t give it context.” Start small, then let the agent pull only the instructions and data it needs for the current task.

Read more

My last open-source skill, /no-ai-slop, went viral with 4K GitHub stars. Today, I’m introducing /human-review, another free AI skill I think you’ll love.

Let’s face it — giving feedback to AI in chat is painful. Asking AI to ‘update the button text’ or ‘edit the headline’ is annoying when I often want to edit the text myself or comment on the exact parts that I want it to fix.

/human-review opens HTML and Markdown files in a visual editor where you can edit text directly, leave comments like a Google Doc, and send feedback to AI.

Get /human-review for free on GitHub

If you find it useful, please star the repository so more people can find it. You can also upgrade to paid to get 12+ more AI skills for personal advice, agent-ready specs, viral social posts, and more.

Watch the demo below, then keep reading to learn:

1. How to install and use /human-review

2. How /human-review works behind the scenes

I’m proud to partner with Oceans Talent

I hired Char through Oceans to help with podcast production. He quickly grew beyond the original role and now creates clips, learns tools like Claude Code and Codex on his own, and delivers more every week.

Oceans Talent finds vetted operators across marketing, operations, finance, and executive support. Their hires know how to work with AI and cost one-third to one-fifth as much as comparable US talent. If you need help, book a call at oceanstalent.com/peter.

Get Started for Free

How to install and use /human-review

To install the skill, paste this into Codex, Claude Code, or your favorite AI coding tool:

After you install it, you can open the visual editor by typing:

In the visual editor, you’ll see a preview of your HTML or Markdown file on the left and a feedback panel on the right:

Now you can:

1. Edit text directly and tweak basic formatting (e.g., bold, italic).

2. Resize images by dragging their corner.

3. Give AI feedback by selecting text or a component and leaving a comment.

4. Send edits and comments in one batch by clicking the “Send X to agent” button.

Giving feedback directly inside an HTML or Markdown file instead of in a separate chat window is super useful. I use /human-review to:

1. Edit planning docs before building new features.

2. Edit and give feedback on landing pages.

3. Tweak product copy and remove AI slop from the UX.

You can also Command-click links in the visual editor to navigate between pages and leave feedback on each one.

How /human-review works behind the scenes

When you run /human-review, it starts a small local server that records your changes in the visual editor and passes them back to your AI agent.

1. Every edit, deletion, and comment is saved and grouped by page.

2. Your agent waits for feedback in 10-minute windows and receives everything immediately when you click Send.

3. Note: For HTML files, direct edits and resizes save automatically. For Markdown and localhost pages, click Send so your agent can apply them to the source.

Because the whole review loop runs locally, /human-review doesn’t require an account or API key and doesn’t upload your feedback to a separate cloud service.

Run /human-review to prevent AI from turning your project into slop

It’s easy to give AI feedback in chat and trust it to edit your documents without reviewing the actual files.

Over time, I’ve found that relying on AI to make edits can slop-ify your entire project if you’re not careful.

So the next time your agent creates an HTML or Markdown file:

1. Type “/human-review (your file)” to open it in the visual editor.

2. Make a direct edit, remove slop, or leave a comment for AI to fix.

3. Hit Send and let your agent update the source.

Get /human-review for free on GitHub and star the repo if you find it useful!

P.S. Credit to my friend Kun, whose tool Lavish inspired /human-review. I built my own version to add direct editing and Markdown support.

Today, I want to share a new episode with Karan.

Karan is the co-founder of Nous Research, the company behind Hermes, the #1 personal agent and AI app on OpenRouter. We had a great chat about why open-source AI must win and how Hermes gets better the more you use it. Karan also shared a live demo of how he used Hermes to modify his favorite childhood game.

Watch now on YouTube, Apple, and Spotify.

Karan and I talked about:

• (00:00) How Hermes differs from Claude Code and Codex

• (02:20) Why every AI model now sounds exactly the same

• (11:16) How to stop AI from saying “you’re absolutely right”

• (15:16) How Hermes creates skills without bloating its context

• (16:40) The open vs. closed-source debate on AI safety

• (23:40) Using one Kanban board for humans and AI agents

• (26:45) Demo: Rebuilding a childhood Sonic world with Hermes

• (36:07) How two Discord friends started Nous Research

• (42:40) How Hermes became its own biggest contributor

I’m proud to partner with Linear

You can now use Linear Agent to edit documents and project descriptions or set up loops to keep them updated automatically as your code changes. Linear makes agent edits transparent through author attribution and version history.

Try Linear for free

Top 10 takeaways I learned from this episode

How to make Hermes work for you

1. The “personal” in personal agent is not the model. Instead, it’s the agent’s memory of your conversations and the skills that you’ve built with it. Karan barely notices any difference in Hermes’ personality when he swaps models because his agent has accumulated so much context about him. The model is interchangeable, but the context layer is what you should cherish and preserve.

2. Customize Hermes’ soul.md to the personality that you want. After using Hermes for 6+ months now, I’ve customized its soul.md to have:

   • Core truths like be genuinely helpful and avoid sycophancy.

   • Boundaries like respecting my privacy especially in any group chats.

   • The right vibes that make it fun to talk to.

   Here’s my complete soul.md that you can copy and paste into Hermes or another AI agent to improve its personality:

Read more

Upgrade to unlock this live workshop and full access to behindthecraft.com with my best AI courses, skills, tool credits, and workshops.

Get instant access

Here’s the recording and key takeaways from our recent live workshop. Paid subscribers can find all workshop recordings here.

Start with the architecture, not the interface

Read more

Today, I want to share my six-step process to design and build Tastemaker, a website for sharing your favorite movies, TV shows, and games.

You’ll learn how to define the problem, create a design.md, prototype in Claude Design, draft an interactive HTML spec, and build the app with AI.

Watch my 25-minute tutorial to see me cover all six steps live, or keep reading below.

Timestamps:

• (00:00) My six-step AI-native design process

• (00:50) Step 1: Define the user problem before building

• (02:25) Step 2: Get inspired and create a design.md

• (05:00) Step 3: Prototype key screens in Claude Design

• (12:59) Step 4: Create an HTML spec for product, design, and tech

• (17:02) Step 5: Design the core flows, default states, and edge cases

• (20:42) Step 6: Build the app with Claude Code

• (23:07) Recap: Why planning should take at least half your time

I’m proud to partner with Replit

Replit is now available inside Claude Design, so you can design beautiful apps and send them directly to Replit to build. To try it, simply select Share in Claude Design, open the Send to… menu, and choose Replit. Get 3 months of Replit free with code PYANG26 at checkout.

Get Replit for 3 Months Free

1. Define the user problem

I’ve watched thousands of movies and TV shows and played hundreds of video games over the years. But IMDb isn’t a great interface and Letterboxd doesn’t support games.

That’s why I decided to build Tastemaker to share my favorite movies, shows, and games all on one page. To kick off the project, I asked Claude Code to research and define the user problem with this prompt:

Here’s the problem statement Claude came up with:

Note how Claude used my spec skill to find real evidence that other people share the same problem. Of course, if I wanted to turn this into a real business, I would spend more time upfront talking to users to validate the opportunity. But since I’m building this as a personal app, let’s move on to step 2.

Get my /spec skill by upgrading to a paid membership

The last skill that I open-sourced, /no-ai-slop, went viral with 3.1K GitHub stars. I’m just as proud of my /spec skill that I use throughout this tutorial. It creates an HTML spec with the product, design, and tech requirements all in one place. Get it, along with 12+ other hand-crafted skills, by upgrading to a paid membership.

2. Get inspired and create a design.md

A design.md is simply a text file with guidelines for color, typography, spacing, and more. It’s a necessary step to avoid AI generating designs that have that recognizable Claude look.

To create a design.md, I like to start by finding inspiration. Recently, I came across a beautiful app called Monogram and love how it uses color to focus people on the content instead of the UI:

I copied a few of Monogram’s screens into Claude along with this prompt to create the design.md for Tastemaker:

Claude will then create a design.md like this:

For more inspiration, you can also check out Mobbin for app screenshots and designmd.sh for design.md files for popular products.

Quick note: Getting inspiration is different from straight-up copying an app.

I feel comfortable borrowing a little from Monogram’s design because Tastemaker is in a completely different product category.

With the design.md ready, let’s move to Claude Design to prototype the product.

3. Prototype key screens in Claude Design

Claude Design is my favorite AI-native design tool because it feels more like working with a design partner than using a tool.

Tip: You don’t need Fable to prototype with Claude Design. I found Opus perfectly up to the task if you want to save tokens.

Here’s the prompt, along with the design.md, to kick off the Claude Design project:

Read more

Today, I want to share a new episode with Jason Liu.

Jason is a DevEx engineer on the Codex team. In our episode, he showed me how he runs his entire workday using Codex, from his chief of staff thread for Slack and email to setting up long-running tasks that have verifiable goals. Jason wrote the official OpenAI handbook on how to get the most out of Codex, so don’t miss this episode.

Watch now on YouTube, Apple, and Spotify.

Jason and I talked about:

• (00:00) The only job left after AI can do the work

• (02:40) Turn pinned threads into long-running workspaces

• (05:50) Build a chief of staff across Slack, email, and Linear

• (10:12) Teach Codex to write like you for every audience

• (13:33) Mine past sessions for new skills and workflows

• (20:01) How Codex revised a launch video while Jason was out

• (24:23) Build a drum practice site around how you learn

• (25:59) Goal, plan, and worklog for long-running Codex tasks

• (33:52) Why taste starts with knowing what feels wrong

I’m proud to partner with Google AI Studio

Google AI Studio was my go-to prototyping tool when I was a product manager. You can use it to generate three versions of your product, add visuals with Google’s Nano Banana model, and publish them to a custom URL for customer feedback. It’s an easy, free way to move from prompt to something your team or customers can try.

Try Google AI Studio for Free

Top 10 takeaways I learned from this episode

Build a Codex system that improves itself

1. Build a chief of staff Codex thread that manages your email, Slack, and more for you. At 9 am, 1 pm, and 5 pm, Jason’s chief of staff thread reviews his Slack, emails, Twitter DMs (through browser use), and Linear before surfacing priorities with links and draft replies. Here’s a prompt you can modify to set this up:

2. Ask Codex to look at your past sessions to build new skills and workflows. Codex has memory of all your past conversations, so the best way to save time is to ask it to find workflows from your manual iterations that it can build new skills to automate. Here’s a prompt you can paste to do this right now:

Read more

Today, I’m open sourcing my /no-ai-slop skill that removes 20+ patterns of AI slop from any piece of writing.

I use this skill to edit everything from emails to docs, and I think you’ll love it too.

Get the skill for free on Github

If you find it useful, please consider starring the repository so more people can find it and upgrading to paid to get 12+ other quality AI skills I’ve made for personal advisor, agent-ready specs, viral social posts, and more. Keep reading for:

1. How to install and use the /no-ai-slop skill

2. My 25/50/25 rule for editing with AI

3. Why the dark side of slop is always calling

I’m proud to partner with Linear

I’m using Linear to manage my entire podcast production process with my video editors and it’s been great. With Linear MCP, I can create, edit, and track issues directly from Codex and Claude Code.

Try Linear for free

How to install and use the /no-ai-slop skill

To install the skill, just paste this in Codex, Claude Code, or your favorite AI harness:

Then you can use it in three ways:

1. “/no-ai-slop (text)” to edit your writing. The skill will remove all the AI slop patterns and list what it removed.

2. “/no-ai-slop is this slop? (text)” to detect AI writing. The skill will list all the slop patterns that it found.

3. “draft an AI slop post (text)” for some fun. Use it to generate the most cringe slop as satire like my LinkedIn comment below:

Jokes aside, even the best AI models (e.g., Fable 5 and GPT-5.6) still use most of the slop patterns below that /no-ai-slop will remove:

1. Binary contrasts. “It’s not X. It’s Y.”

2. Throat-clearing openers. “Here’s the thing,” “Let me be clear”

3. Faux-insight setups. “What nobody tells you,” “The part everyone misses”

4. Colon reveals. “The best part: it learns.”

5. Dramatic fragments. “That’s it. That’s the whole thing.”

6. Superficial analysis. “highlighting the team’s commitment to innovation”

7. Importance puffery. “marks a pivotal moment,” “a testament to”

8. Weasel attribution. “experts agree,” “studies show”

9. Synonym cycling. “The agent handles your email. The assistant drafts replies.”

10. Fake-profound endings. “The future isn’t coming. It’s already here.”

Here’s an example of using the skill to detect AI slop on a recent tweet I came across:

My 25/50/25 rule for editing with AI

Substack just shipped AI slop detection to protect readers. They featured my video as an example of how to use AI to edit while still sweating the details.

Basically, I work with AI in the middle 50% of my writing and editing process:

1. First 25%: I start by voice dictating or writing a rough draft without AI.

2. Middle 50%: I use my /newsletter and /no-ai-slop skills to work with AI to clean up the draft. AI helps me find grammar mistakes, improve clarity, and more.

3. Last 25%: I manually go through the piece line by line to make edits without AI.

I’d like to think that the difference between a creator who pumps out slop and a creator who cares about quality is giving a shit about the first and last 25%.

This applies to building with AI too. As Kun (AI builder) shared in our interview, AI handles the middle 50% where most of the code gets written. But you have to care enough about the first 25% (using your taste to shape the requirements) and the last 25% (validating the work) to make a product great.

Why the dark side of slop is always calling

The dark side of just letting AI draft and edit everything is always calling:

1. It’s easier to just give AI feedback to edit than to do it manually. Products like Codex and Claude Code seem to actively discourage you from making manual edits to files.

2. It’s easier to just be lazy and not write at all if AI is not available. For example, I used to be productive on long flights. But lately, I’ve become too lazy to write or edit anything without having AI available.

Perhaps most depressing of all, pumping out slop can drive real results.

Without naming names, I’ve seen some creators publish 10 AI slop posts per hour, every day. Some of those posts go viral and bring them more followers and revenue. But I like to remind myself that:

People who know AI can see the slop coming from a mile away.

A post can perform while still costing you trust with the people whose judgment matters to you. Consider these two examples:

1. Brian Chesky’s account got hacked and posted an obviously AI generated thread. Readers spotted it as AI before anyone confirmed the hack.

   

2. My friend, Alex Cohen, wrote a letter to his employees highlighting that pumping out slop puts an unfair burden on the reader.

Don’t let AI make you dumber and lazier

I’m not going back to editing everything manually, but I also don’t want to lose my ability to think or craft a sentence when there’s no Wi-Fi available.

So I’m going to hold the line at my 25/50/25 writing process. Trust is the only durable asset we all have, and slop hurts trust with the people who matter.

Get my free /no-ai-slop skill on Github and let me know what you think!

Get instant access

Here’s the recording and key takeaways from our recent live workshop. Paid subscribers can find all workshop recordings here.

How Julie’s AI-native design workflow works

Read more

Today, I want to share a new episode with Thariq Shihipar.

Thariq works on the Claude Code team, and I’ve wanted to see how he builds for a long time. In our episode, he showed how to use /goal to keep Claude working, how he plans with Claude to remove unknowns before building, and how he runs a team of agents in Slack. He also shared why his team cut Claude Code’s system prompt by 80% and how he creates polished launch videos with Claude.

Watch now on YouTube, Apple, and Spotify.

Thariq and I talked about:

• (00:00) Goals, loops, and workflows to make Claude work longer

• (04:48) Live demo: Creating a polished video with Claude

• (08:17) Plan by removing unknowns before you build

• (14:32) How Anthropic uses HTML artifacts for planning

• (18:35) Running parallel Claude tasks through Slack

• (27:34) Using separate creator and verifier sub-agents

• (31:08) How Thariq stays focused while Claude works in parallel

• (34:07) Why Anthropic cut Claude Code’s system prompt by 80%

• (37:01) How to become more technical when AI writes the code

I’m proud to partner with Wispr Flow

When I’m building with Claude Code, I use Wispr Flow to dictate all my prompts instead of typing. It’s more accurate than other AI voice dictation tools and saves me hours every week. Try it out for free with my exclusive link below.

Try Wispr Flow for Free

Top 10 takeaways I learned from this episode

1. When a better AI model is released, the first thing you should try is to remove instructions. Thariq and the Claude Code team cut Claude Code’s system prompt by 80% because hard rules and examples were constraining newer models. When a new model is released, audit your CLAUDE.md and skills to remove deterministic instructions.

2. Plan with Claude to discover and remove unknowns. Before building his one-shot video workflow, Thariq asked Claude to explain how voice transcription tools like Whisper could fail. This helped him clarify the plan before getting Claude to create the video. Perhaps the most important learning here is that he actually reads the plan manually instead of just trusting AI.

3. Use /goal only if you have a verifiable finish line. I asked Claude to “/goal Build me a great game” and the results were…not great. Thariq shared how to use /goal to create polished videos with Claude by providing a verifiable end state (e.g., a rendered video). Here’s the exact prompt he used to create the video in one shot:

Read more

Get instant access

Dear subscribers,

I now use ChatGPT Work and GPT-5.6 for almost everything I do on my computer, and I’m going to show you my full setup in this video and written guide.

You’ll learn how to choose the right GPT-5.6 model, connect your apps, manage email and calendar, automate recurring work, build Sites, and more.

Watch my tutorial now for the full 28 min walkthrough, or keep reading below.

Timestamps:

• (00:00) How I use ChatGPT Work for almost everything

• (01:04) Chat vs. Work vs. Codex: What’s the difference?

• (02:22) Which GPT-5.6 model and effort to use

• (03:18) Personalizing ChatGPT Work’s custom instructions

• (06:23) Connect your apps and organize your Work threads

• (09:30) Let ChatGPT manage email and draft replies in your voice

• (13:47) Schedule events and prep for meetings automatically

• (17:19) Automate recurring work with scheduled tasks

• (19:23) Continue your Work tasks from your phone

• (21:01) Publish travel guides, games, and product plans as Sites

I’m proud to partner with Wispr Flow

I use Wispr Flow to dictate first drafts much faster than I can type. It helps me get my ideas down while they’re still fresh, so I can focus on editing. I use it all the time in ChatGPT Work and Codex, and it’s been a huge time saver. Try it for free with my link.

Try WisprFlow for Free

Chat vs. Work vs. Codex: What’s the difference?

Here’s the simplest way to think about the three modes:

1. Chat is the regular ChatGPT we’re all familiar with.

2. Work lets ChatGPT take action in your favorite apps.

3. Codex adds features designed for coding and developers.

Honestly, I’m not a fan of Work and Codex existing as separate modes. As far as I can tell, there are only small differences in the UI and how responses are presented. For example, Codex adds a Pull Request tab that developers can use to review each other’s code.

So I don’t think it makes much difference whether you use Work or Codex. My recommendation is simple:

Use the ChatGPT desktop app which is far more capable than ChatGPT for the web.

Which GPT-5.6 model and effort should you use?

GPT-5.6 comes in three variants (Sol, Terra, and Luna), each with five effort levels from Low to Ultra. So which model and effort should you default to?

Here’s a chart that shows intelligence vs. output tokens for the various options:

Based on this chart and my experience, I recommend that you:

Use GPT-5.6 Sol (medium) for the best balance of intelligence and token use.

I sometimes use Sol (high) for planning, but overall I haven’t noticed much of a difference between that and medium. I almost never use Terra or Luna.

Personalize your ChatGPT Custom Instructions

On a fresh install, I go to Settings → Personalization and set my Custom Instructions to:

1. Teach me technical concepts without using jargon.

2. Challenge my thinking instead of agreeing with every prompt.

3. Remove AI slop phrases and patterns.

Here are my exact custom instructions if you want to copy my setup:

Install plugins to let ChatGPT use your favorite apps

Plugins connect ChatGPT Work and Codex to the apps you use every day.

The screenshot above shows some of the plugins I have installed. I think the bare minimum set is:

• Gmail

• Google Calendar

• Documents, Presentations, Spreadsheets, and Google Drive

Other useful plugins include the PDF plugin for editing PDFs, Slack for messages, Notion for wiki articles, Figma for design, and Granola for meeting transcripts.

Install plugins for any app that you want ChatGPT to use. This is the key setup step that makes ChatGPT Work and Codex your everything app.

Use long-running threads for each type of work

As you use ChatGPT Work more, you may get overwhelmed by all your active projects and threads. To avoid this:

Keep one long-running thread for each type of work.

For example, in my Personal OS project, my threads include:

1. Newsletter to edit all newsletter posts.

2. Social for drafting and making posts for social networks.

3. Podcast to prep and do post-production for interviews.

ChatGPT Work and Codex are very good at compacting long conversations, so you can keep chatting in the same thread without worrying about GPT getting confused.

I also pin my most-used threads, such as my Advisor thread for strategy and business advice and my Chief of Staff thread for email, calendar, and more.

Set up a chief of staff thread for email and calendar

These days, I manage my email and calendar almost entirely through ChatGPT Work and Codex instead of opening Gmail or Google Calendar. Let me walk through how I:

1. Find forgotten follow-ups and subscriptions to cancel

2. Draft email replies that sound like you

3. Manage your calendar and avoid meeting conflicts

4. Get briefed on upcoming meetings

1. Find forgotten follow-ups and subscriptions to cancel

Read more

Get instant access

Dear subscribers,

Today, I want to share a new episode with Jared Zoneraich.

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.

Watch now on YouTube, Apple, and Spotify.

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

I’m proud to partner with Riverside

I produce a podcast every week and it used to take me at least 10 hours to record, edit, cut clips, write posts, and publish everywhere. Riverside has cut that time by half.

In addition to recording at 4K local with clean audio, I can also now edit using the transcript, add captions, make clips, and create and publish social posts all from one place. Riverside is trusted by teams like Spotify, Microsoft, and Vercel and you can try it for one month free at creators.riverside.com/PeterYang with code PeterYang.

Try Riverside for Free

Top 10 takeaways I learned from this episode

How to build great AI agents

1. Start by expressing your intent to the model in a simple prompt. “Let the model cook” by expressing your intent and constraints clearly instead of forcing it to follow deterministic instructions.

2. Give the model relevant tools to pull context. “Tool engineering” is giving the model ways to search your code, read logs, run tests, open the app, and inspect what it changed.

Read more