"Launch and iterate" was the startup mantra for years. A team would build v0, launch it, get feedback from users, and then iterate based on this feedback to launch v1. The key message was to launch sooner rather than later so you could begin getting valuable feedback from customers.

This approach emerged because adding features was expensive, and it was easy to get sucked into a cycle of feature bloat where v0 would take a long time to ship. When it finally did ship after several months or even quarters, the team might find that it missed the mark and users didn't find the product valuable. This would be a costly mistake.

In the age of personal software, where you can build applications for yourself using AI, I find that "launch and iterate" has been replaced by "prompt and iterate." You prompt AI to build software to solve a specific problem that you have. Then you immediately test it (since you're the only user, you only need your feedback and it's instantaneous). Then you prompt AI to make changes to your software based on your testing. In this loop, the "launch" itself is almost a non-event. You're not spending any time on that at all. You're spending time instead on prompting and testing.

Building ScribbleTab

To show how this works in practice, let me walk you through building ScribbleTab, a Chrome extension to draw, highlight, and comment on any web page. There are existing Chrome extensions that do this, but I decided to build one tailored to my requirements. For example, I wanted to be able to move the toolbar, toggle its visibility using keyboard shortcuts, and implement undo/redo functionality.

Here's how I built it using the "prompt and iterate" cycle. Since the Claude 4 models dropped recently, I used this opportunity to test the Opus 4 and Sonnet 4 models. I used the larger Opus model to write the spec for the extension and then used Sonnet to implement it.

Starting with the Spec

My first prompt to Opus was straightforward:

Create a Chrome extension that will let me draw on the screen, select/annotate shapes to showcase UI elements, draw text, etc. It should work on a layer above what's on the screen. Use Manifest V3.

Write the spec for this, but do not code.

Here's the detailed spec that Opus wrote:

Chrome Extension Specification: Web Annotator

Overview

A Chrome extension that provides an overlay drawing and annotation layer on any web page, allowing users to draw, highlight UI elements, add text annotations, and create visual guides directly on top of web content.

Core Features

Drawing Tools

Freehand Drawing: Pen/pencil tool with variable thickness and color options

Shapes: Rectangle, circle, arrow, line tools for structured annotations

Highlighter: Semi-transparent marker for emphasizing content

Eraser: Remove individual annotations or clear all

Text Annotations

Text Boxes: Click to place resizable text boxes anywhere on screen

Font Options: Size, color, basic formatting (bold, italic)

Callout Bubbles: Speech bubble-style annotations with connecting arrows

UI Element Selection

Smart Selection: Hover to highlight DOM elements with visual feedback

Element Outlining: Draw attention boxes around selected elements

Element Information: Optional tooltip showing element type, class, or ID

Layer Management

Overlay Canvas: Transparent layer that sits above all page content

Z-index Management: Ensure annotations stay on top of page elements

Scroll Synchronization: Annotations move with page scroll

Viewport Persistence: Maintain relative positions during window resize

Controls & Interface

Floating Toolbar: Draggable control panel with tool selection

Keyboard Shortcuts: Quick access to common tools (e.g., Ctrl+Z for undo)

Color Picker: Preset colors plus custom color selection

Tool Options Panel: Context-sensitive settings for selected tool

Technical Architecture

Manifest V3 Structure

Service Worker: Background script for extension lifecycle management

Content Script: Injected into web pages to create annotation layer

Action Popup: Quick access menu for enabling/disabling annotations

Options Page: Detailed settings and preferences

Permissions Required

activeTab: Access to current tab for content script injection

storage: Save user preferences and annotation data

scripting: Inject content scripts dynamically

Data Storage

Local Storage: User preferences, tool settings, color history

Session Storage: Temporary annotation data per tab

Export Options: Save annotations as image (PNG) or data (JSON)

User Workflow

Activation

Click extension icon or use keyboard shortcut

Overlay appears with floating toolbar

Page content remains interactive below annotation layer

Drawing Process

Select tool from toolbar

Configure tool options (color, size, etc.)

Click/drag on page to create annotations

Use selection tool to modify existing annotations

Deactivation

Click extension icon again or use escape key

Option to save or discard current annotations

Overlay disappears, returning normal page interaction

The Iteration Process

I was happy with the spec Opus wrote, so I switched to Sonnet 4 for the actual implementation. Within 10 minutes, I had a v0 of the extension installed in my browser. Then I began a series of test→prompt→test cycles that looked like this:

Iteration 1: Claude's initial implementation worked well for drawing tools, but text annotations weren't working. Also, I realized I could not minimize or hide the toolbar and it was in my way. I told Claude the text tool was broken and that I needed to hide the toolbar without losing annotations. Within minutes, Claude fixed the text positioning logic and made it possible to toggle the toolbar's visibility.

Iteration 2: After more testing on different web pages, I realized the toolbar was stuck in one position and just being able to toggle its visibility wasn't flexible enough. I wanted to make the toolbar movable. I asked Claude to implement this and soon I had this working flawlessly.

Iteration 3: Further testing revealed that the undo system was broken. Clicking Undo would delete all annotations. I prompted Claude to fix this and also asked for redo functionality to be added.

Iteration 4: My extension was "feature complete" at this point, so I decided to improve its look and feel. Claude had used various emojis for toolbar icons and these looked hideous, so I bluntly told Claude the icons sucked and asked it to replace every icon with crisp, professional SVG designs.

Iteration 5: Claude replaced emojis with beautiful new SVG icons. However, this change caused tool selection to become janky and unreliable. I turned on extended thinking mode and asked Claude to fix this issue while keeping the new icons.

The Result

I finally arrived at a product I was happy with. The extension did exactly what I needed: it allows me to quickly annotate web pages with drawings and text, has the keyboard shortcuts I wanted, and a draggable toolbar that stays out of my way.

You can view the code on my GitHub.

I even created a banner image and an icon for Scribble Tab using ChatGPT's image generation feature.

Reflections on Prompt and Iterate

This experience once again reinforced how fundamentally different building personal software has become. When you're building for yourself, you already know exactly what you want, you just need the time and skills to build it.

The prompt-and-iterate approach lets you focus entirely on the problem-solving and refinement process, rather than the mechanics of coding. Each iteration cycle was minutes rather than hours, and I could test immediately since I was the only user that mattered. It's the significant advancements in the coding capabilities of AI models like Claude 4 and Gemini 2.5 pro that make this rapid, personal development cycle a reality.

In total, it took me just over 2 hours from start to finish. Without the help of AI, it would have easily taken me weeks to build this extension from scratch. In fact, I would not even have attempted to build it because I couldn't have afforded to spend so much time on it. That's the other huge benefit of AI: it makes the impossible suddenly feasible.

Have you used AI for personal projects in a similar way? What are your experiences with 'prompt and iterate'?

Let me tell you about a few ways my habits have already started to shift.

Learning Things Way Faster

Remember trying to understand something really complicated and just feeling stuck? I used to spend ages trying to figure things out. Now, if I come across a dense research paper, I can ask an AI to break down the main ideas in simple terms. It’s like having a super-smart tutor on call.

And it’s not just for reading. If I’m struggling to really get a concept, I can ask AI to help me picture it, or even sketch out how a little interactive demo would work. For example, I wanted to understand regular expressions better. So, I had Gemini help me design a simple website where I could try out different patterns and see them work instantly. It’s much better than just reading about it; I can play around until it makes sense. This makes learning feel more like a hands-on exploration.

Building My Own Tools (It's Easier Than You Think!)

I used to always search for an app someone else had made whenever I had a specific need. But now, I’m finding I can often build my own little tools, especially with AI helping out with some of the coding.

For instance, I often need to share screenshots, but some information needs to be blurred out for privacy. Instead of fiddling with editing software every time, I built a simple Chrome extension that lets me quickly blur parts of my screen before I take the shot. It solved the problem of needing to quickly obscure sensitive details, and it really saves me time. Plus, since it’s my own extension, I know exactly what it does and don’t have to worry about the security or privacy risks that can come with installing third-party browser tools.

Another little app I made, which is also a Chrome extension, helps me bookmark webpages directly into a Google Sheet. This is perfect for organizing links I want to save because all the information is stored in my own spreadsheet. This means it’s private, and I can easily open it up anytime to find links I’ve previously saved. I’ve even put together small scripts, like one that emails me my upcoming calendar events each week, or a utility that copies code from a folder right to my clipboard so I can easily paste it into LLMs to enhance it.

It really feels like we're entering an age of "personal software." Instead of a one-size-fits-all solution, I can make something that fits my exact needs, and then change it as I go. It's pretty cool to have that kind of control.

Getting More from My Books

I’ve always loved books and have a big collection on my Kindle. But lately, I’ve found myself preferring to buy books as PDF files when I can, especially ones that aren't locked down by DRM. The main reason? I can load these PDFs into tools like Google's NotebookLM.

Imagine reading a book about Big Data and being able to chat with it. I can ask things like, "What are the key takeaways from chapter 3?" or "Can you explain the author's viewpoint on Collaborative Filtering in simpler terms?" It’s like having an intelligent conversation partner embedded in my reading material. This helps me engage with the content much more deeply and remember it better than just highlighting text. It makes my books feel much more dynamic.

Sharing Memories, Keeping Our Privacy

My wife and I tend to be private people, especially when it comes to posting pictures of our family and kids online. We want to share our joys, but we're also careful about privacy in the digital world.

AI image generation has offered a neat solution here. If we have a photo from a special family occasion, like the kids building a sandcastle on a beach, I can use AI to create an artistic interpretation of that moment. It could be in a storybook illustration style or look like a soft painting. The AI image captures the feeling and the essence of the memory – you can still tell it’s our family and what’s happening – but it’s not their exact photographic likenesses. This way, we can share the happiness with friends and family without compromising our kids' privacy. It’s a good balance for us.

The New Normal?

These are just some of the ways AI is starting to weave into my daily life. It’s less about AI taking over and more about it providing tools that allow me to do things in new, often better, ways – learning more effectively, creating personalized solutions, and sharing on my own terms.

I’m curious if you’re seeing similar changes. Are you using AI in ways that surprise you, or that have made your daily tasks a bit different? I’d love to hear your experiences in the comments below!

Checklists are simple to use and have helped me successfully complete tasks with confidence. Using checklists has made me more productive and efficient while reducing stress and anxiety. I encourage everyone to try using checklists to experience these benefits themselves.

What is a checklist?

A checklist is simply a set of steps that you need to complete in order to accomplish a task. These tasks could be "Packing for a road trip" or "Filing tax returns", etc. I have found that checklists are especially useful for tasks that are repetitive or that have a lot of steps.

The Power of Checklists

I didn't fully appreciate the value of checklists when I read "The Checklist Manifesto." In that book, the author makes a compelling argument for checklists by describing how they have significantly reduced human error in high-stakes environments such as aviation and medical surgeries. At that time, I was already using lists to track TODOs, but the concept of a checklist was new to me. It wasn't until my costly Roth IRA mistake that I truly understood the power of checklists to prevent mistakes.

The key difference between a checklist and a TODO list is that the former focuses on making a specific task successful, whereas the latter tracks various tasks that you need to complete.

When to create a checklist

After seeing how a simple checklist could have prevented such an expensive mistake, I began seeing opportunities for checklists everywhere.

I began creating checklists in two types of scenarios: (1) I reactively created them after forgetting an important step while performing a task. This ensured that I didn't repeat the same mistake(s) the next time I performed the task. (2) I proactively created them before performing high-stakes tasks.

The first checklist I created was a "Personal finance" checklist, motivated by that initial Roth IRA mistake. Since then, I have reactively created checklists for various tasks such as:

• Packing for various kinds of trips

  • Road trips

  • Domestic trips

  • International trips

• Day of trip

In all of the above cases, it was a failure of some sort that prompted the creation of a checklist for that particular task. For example, my first packing checklist was prompted by a trip where I had forgotten to pack medicines, sunscreen, slippers, and even my travel toothbrush. I had actually packed in advance, so my error wasn't due to a last-minute scramble. I had simply forgotten to pack these items. I had even taken the time to run through a list of things to pack in my mind, but relying on my memory alone was clearly not sufficient. Since creating a packing checklist, I have never once forgotten to pack items on the list.

As another example, I created a separate "Day of trip" checklist when I forgot to take the trash out before heading out on a trip.

I have also proactively created checklists for important tasks like "Tax filing" where it is important to get it right the first time.

Remember to update checklists

Checklists aren't meant to be static. It is important to keep these updated over time as your needs and circumstances change. I review my checklists after each use and ask myself: "Did I miss anything?" or "Were any steps unnecessary?" For example, my travel checklist evolved significantly after having kids. Similarly, my personal finance checklist has grown to include new investment accounts and updated contribution limits. Regular review ensures your checklists remain relevant and continue to serve their purpose effectively.

Why checklists work for me

Checklists help me systematize and codify my knowledge and experience by offloading key details from memory. Instead of spending mental energy remembering every step of a complex process, I can focus on executing each step well. This reduces cognitive load and decision fatigue, especially for tasks I don't perform frequently. For instance, when doing my taxes, I don't have to worry about forgetting to report a particular type of income or missing a deduction - it's all captured in my checklist. This mental offloading also helps reduce anxiety about forgetting important steps, allowing me to perform tasks with greater confidence and clarity.

Tools for creating checklists

There are many different ways to create a checklist. I primarily use Google Sheets for complex checklists that require multiple sections or additional notes. For example, my tax filing checklist in Google Sheets has separate sections for gathering documents and reporting different types of income. Each item includes notes about where to find specific information or what to watch out for. For simpler checklists, I use Google Tasks because it's easily accessible on my phone and allows me to quickly check off items as I complete them. While there are specialized checklist apps available, I prefer these basic tools because they're free, reliable, and integrate well with my existing workflow. The most important thing is to find a method that works for you and that you will use consistently - whether that's a paper notebook, a spreadsheet, or a dedicated app.

When you're both the developer and the sole user, you're guaranteed to have a perfect understanding of the requirements. Don't like a feature? Change it. Need something specific? Add it. The feedback loop is instantaneous, and the iterations are driven purely by your needs.

Today's technology landscape offers numerous entry points for personal software development. Chrome Extensions can enhance your browsing experience, automating repetitive tasks or adding features you've always wished existed. Google Apps Script lets you create powerful automation within your workspace. Simple mobile apps can be built using platforms App Sheet, while personal blogs give you control over your online presence.

It's never been easier to build these simple apps thanks to recent technological advances. Large Language Models (LLMs) like Claude Sonnet 3.5 and Gemini Flash 2.0 can assist with coding and debugging, serverless platforms eliminate infrastructure headaches, and low-code/no-code tools provide visual development environments. You don't need to be a seasoned developer to start building – you just need a problem worth solving for yourself.

There are several advantages to personal software development. You gain complete control over your digital tools, with the flexibility to modify them as your needs evolve. Customization reaches levels impossible with commercial software, as every feature is crafted to your specifications. Building personal tools can help you learn new technical skills. Cost-wise, while there's an investment of time, you're free from recurring subscription fees. Perhaps most importantly, there's an immense sense of pride and joy in using tools you've created yourself.

However, even when building for yourself, it is important to prioritize privacy and security. These personal projects shouldn't compromise your data security or privacy. Furthermore, sometimes the right decision will be to buy software versus build it yourself. When the stakes are high or the effort to build it is high, it may make more sense to buy the software instead of building it yourself.

The next time you find yourself frustrated with existing software options, remember: you have the power to build exactly what you need. Your perfect tool is waiting to be created – by you, for you. Start simple, focus on solving real problems you encounter, and gradually expand your personal software ecosystem as your skills and needs grow.