Prompts, patterns, and dispatches from the frontier of AI-augmented software engineering.https://ralphloopsarecool.com/https://ralphloopsarecool.com/blog/autonomous-agents-are-overhyped/https://ralphloopsarecool.com/blog/autonomous-agents-are-overhyped/AI tool companies want you to believe agents can replace developers. They can't. Here's why the human-in-the-loop isn't a limitation — it's the entire point.Wed, 18 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/claude-opus-4-review/https://ralphloopsarecool.com/blog/claude-opus-4-review/Opus 4.6 isn't flashy. It doesn't generate images or browse the web. But for actual productivity work, nothing else comes close.Sat, 14 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/spec-driven-development/https://ralphloopsarecool.com/blog/spec-driven-development/Vibe coding is fun until your agent forgets what it is building. Specs fix that. Here are the best practices for making them work.Sat, 14 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/how-generative-ai-works/https://ralphloopsarecool.com/blog/how-generative-ai-works/LLMs aren't magic and they aren't thinking. Once you understand what's really happening under the hood, you stop fighting them and start getting results.Tue, 17 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/everything-is-a-loop/https://ralphloopsarecool.com/blog/everything-is-a-loop/From stacking features to spinning feedback loops: how I changed the way I build software.Tue, 10 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/the-pink-elephant-problem/https://ralphloopsarecool.com/blog/the-pink-elephant-problem/Why telling an AI what to avoid makes it more likely to do exactly that — and what to do instead.Thu, 12 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/agents-md-outperforms-skills/https://ralphloopsarecool.com/blog/agents-md-outperforms-skills/Vercel found that passive context in AGENTS.md beat active skill retrieval — hitting 100% pass rates where skills stalled at 53%. Here is why that matters.Sun, 15 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/mac-mini-agent-server/https://ralphloopsarecool.com/blog/mac-mini-agent-server/A dedicated Mac Mini running your AI agents keeps experiments off your daily driver, gives you persistent context, and costs less than you'd think.Sun, 15 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/behavior-driven-prompting/https://ralphloopsarecool.com/blog/behavior-driven-prompting/The most effective AI coding workflow starts with a PRD, decomposes into behaviors, breaks those into tasks, and maintains a living spec of what the system actually does.Fri, 13 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/writing-better-prompts/https://ralphloopsarecool.com/blog/writing-better-prompts/A practical guide to writing prompts that actually work — starting with the one mistake almost everyone makes.Wed, 11 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/clawdbot-setup-guide/https://ralphloopsarecool.com/blog/clawdbot-setup-guide/OpenClaw (formerly Clawdbot) is a self-hosted AI agent that connects Claude, GPT, and Gemini to your messaging apps and desktop. Here's how to set it up on a Mac Mini.Mon, 16 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/comparing-autonomous-agents/https://ralphloopsarecool.com/blog/comparing-autonomous-agents/Four autonomous coding agents, one honest look at the 2026 data — 1.7x more bugs, masked failures, infinite loops, and what actually separates the tools that work from the ones that don't.Thu, 19 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/llm-training-deep-dive/https://ralphloopsarecool.com/blog/llm-training-deep-dive/Going deeper than next-word prediction — what embeddings, attention, and RLHF actually mean, and why understanding the training process makes you dramatically more effective with these tools.Sat, 21 Feb 2026 08:00:00 GMThttps://ralphloopsarecool.com/blog/comment-driven-development/https://ralphloopsarecool.com/blog/comment-driven-development/Spec-driven development is a real workflow with a product attached. Comment-driven development is the older, simpler version — and it works even better with agents.Tue, 24 Mar 2026 07:00:00 GMThttps://ralphloopsarecool.com/blog/night-shift-agentic-workflow/https://ralphloopsarecool.com/blog/night-shift-agentic-workflow/Human attention is scarce. Agent compute is cheap. The best workflows treat these as different shifts, not the same job.Fri, 20 Mar 2026 07:00:00 GMT