Software Engineer & AI Enthusiast
The first version of agent-config tried to impose uniform rules on Claude, Codex, Copilot, and Gemini — and it broke. The fix wasn't better engineering up front. It was accepting failure fast, letting AI handle per-agent format conversion, and shipping a v2 that actually works. Here's what the failure taught about human instinct, AI-assisted iteration, and why failing cheaper is a skill worth building.
Between Claude Pro, ChatGPT Plus, GitHub Copilot, Cursor, and a handful of specialty tools, a serious AI stack can run $100 or more per month. This post explores a practical closed-loop strategy: give every subscription a job, pick one small revenue experiment, and use the tools themselves to generate enough income to cover the bill.
As AI features proliferate across SaaS products, customers are waking up to a frustrating reality: they're paying for the same AI capabilities multiple times. The Bring Your Own Key (BYOK) model — letting users connect their own Anthropic, OpenAI, or Gemini API keys — is the market's answer. This post unpacks the real tradeoffs on both sides and maps out which model fits which kind of user.
Attractor is a DOT-based AI pipeline orchestration engine that borrows its name and philosophy from dynamical systems theory. Define your workflow as a Graphviz directed graph, and Attractor pulls multi-LLM execution toward your desired outcome — through conditional branches, retries, human gates, and failure recovery. The entire codebase is AI-generated, making it a live proof of the Software Factory paradigm it embodies.
Understanding how large language models actually work under the hood removes some of the wonder — but it also replaces irrational fear with informed clarity. AI isn't magic, it isn't plotting against us, and knowing why might just change how you think about your own brain.
Read more
Discover how to manage Claude, Codex, and GitHub Copilot configurations from a single source of truth. The coreydaley/agent-config GitHub repository provides a centralized home for agent instructions, reusable skills, custom commands, subagents, and prompts—with a simple Makefile-based setup that symlinks everything into place across your AI tools.
Accessibility is something every website owner should care about, yet it's easy to deprioritize. I asked Claude Code to audit my Hugo blog for screen reader compatibility and was amazed at both the depth of its findings and how quickly it fixed every issue — from missing skip-navigation links to unlabeled SVG icons to dynamic search results that screen readers couldn't detect.
Read more
There's a special kind of frustration that comes from watching a product you love ship a hidden dinosaur game or a secret disco mode while a bug you reported two years ago collects digital dust. But is that frustration always warranted? Let's unpack why companies build these things, who actually builds them, and whether they really steal time from the work that matters.
Read more
As AI coding agents become mainstream, the traditional coding interview is evolving. Explore how live AI-assisted problem-solving, collaborative workflows with planning modes, and practical project management may replace LeetCode-style assessments, while examining the ethical implications of this transformation.
Read more
Announcing the release of coreydaley-dev, a fun and functional Hugo theme now available on GitHub under the MIT license. This post explores why I'm open sourcing the theme, its standout features including special date avatar swapping, the collaborative AI-assisted development process, and how you can use it for your own projects. Discover a Hugo theme that balances personality with performance, built through human-AI collaboration and designed to make static sites more engaging and distinctive.
Read more