How I Use the Reflection Pattern in My Day-to-Day AI-Assisted Coding
A practical look at using ChatGPT, Claude, and Cursor as a multi-agent reflection loop — and when to actually use it.
Notes
Random thoughts, tutorials, and things I learn along the way.
I keep these intentionally short to deliver the point — I dislike long posts. ¯\_(ツ)_/¯
Note: Some of this content was written with AI assistance, but the thoughts, experiences, and insights shared here are genuinely mine. My goal is simply to share what I've learned along the way.
AI, Privacy, and the Case for On-Device Language Models
Exploring the privacy benefits and technical tradeoffs of running AI models locally on user devices instead of in the cloud.
AI Agents DIY Project: A Custom Connector for ChatGPT to Control Claude-Code Sessions
How I built a DIY MCP server to let ChatGPT remotely orchestrate Claude-Code sessions — enabling real coding work while away from my desk.
Why Reading the Chain of Thought Matters More Than the Code (Sometimes)
A reflection on how staying inside the reasoning loop — instead of passively waiting for results — makes AI coding more efficient for big changes and critical fixes.
Dead Ends, Agent Loops, and the Case for Human-Centered Failures
Why product teams should design for failure, not just the happy path—lessons from agent loops and a frustrating Uber Eats experience.
Recursive GPT Building: Letting the Bot Be Builder, Tester, and Critic
How I used the Custom GPT builder itself to refine, test, and critique a new GPT — while keeping myself in the loop where it mattered.
Custom GPTs That Test Themselves
How I use self-simulated user testing to make custom GPTs smoother, faster, and more fun to use.
Avoiding Chaos: The Right Team Setup for Agentic AI Products
For fast-moving GenAI teams, structure matters. Here's how to separate iteration from engineering — and ship smarter.
Human Memory, Struggle & AI-Assisted Coding
How AI tools can actually make our memory worse, and the personal strategies I use to stay cognitively engaged while coding.
GenAI MVPs: The Real Work Begins After You Release
Hard-earned lessons from building and shipping fast in the world of AI assistants and chatflows.
Track Your Focused Work: Break Free from Stress and Boost Productivity
Learn how working memory, stress, and distractions impact productivity, and discover strategies to improve focus and reduce workplace stress for better performance.
The Stress of Stress is Worse than Stress: How to Build Better Coping Mechanisms
Learn practical strategies for managing workplace stress and building effective coping mechanisms based on a decade of experience in the software industry.
Forget Goals. Build Systems.
Why systems beat goals, and the one book that survived my productivity purge.
How to Bypass Vercel's Free-Tier Deployment Limits Using GitHub Actions
Learn a practical way to manage deployments beyond Vercel's free tier limitations using GitHub Actions.
Overkilling Code Reviews
Learn how to avoid over-engineering your code review processes and build contextual review systems that match your team's actual needs rather than following FAANG company practices.
Create A 'Control Room' for Client-Side Rendered Apps
Learn how to build a control room system for client-side rendered applications using WebSockets to manage app state, maintenance windows, and version updates in real-time.
Create Editable HTML in React
Learn how to build editable HTML components in React using contentEditable, with a generic solution for inline editing across your application.
Custom React Hook to Share State Between Browser's Windows
Learn how to build a custom React hook that enables state synchronization across multiple browser tabs using localStorage and the storage event API.