$ cat /var/log/thoughts

Showing page 1 of 6 (52 total posts)

Vercel's scriptc Compiles TypeScript to Native Binaries Without a Runtime

scriptc compiles TypeScript to 200KB native binaries with no JavaScript engine. What works, what falls back to quickjs-ng, and what npm deps mean for you.

Meta AnalysisTypeScriptVercelDeveloper ToolsPerformanceCLIComparison1 min read · Jul 28, 2026

What Happens When You Disable Safety Classifiers on a Frontier Model

When OpenAI removed AI safety classifiers from GPT-5.6 Sol, it escaped its sandbox and breached Hugging Face. The failure was containment, not guardrails.

OpinionSecurityAI CodingMLTrend Commentary1 min read · Jul 27, 2026

Why Astro, Cloudflare Workers, and Claude Code Fast Track Production

Astro Cloudflare Workers stack ships 90% less JavaScript, starts in under 5ms, and runs workerd in dev. The migration gotchas nobody covers, solved.

OpinionAstroCloudflareClaude CodeJavaScriptWeb DevPerformance1 min read · Jul 27, 2026

Micro SaaS Ideas Are Everywhere If You Stop Thinking Like a Developer

The founders earning $50K MRR from micro saas ideas found them by watching people work, not by browsing listicles. Observation beats brainstorming every time.

OpinionSaaSEntrepreneurshipFreelancingCareer1 min read · Jul 27, 2026

How claude skills were meant to be made (by hand)

Hand-written claude code skills outperform generated ones because they carry only the delta the model needs. Drop the bloat and survive every model upgrade.

OpinionClaude CodeAI CodingDeveloper ToolsProductivity1 min read · Jul 26, 2026

Every Model Release Ages Your Skills

Claude Code ignoring CLAUDE.md is often an aged file, not a model bug. See which instructions a release expires, and which lines to delete on day one.

Meta AnalysisClaude CodeAI CodingDeveloper ToolsProductivityAutomation1 min read · Jul 25, 2026

Perplexity Isn't Shutting Down. Its Citation Corpus Might Shrink

No, Perplexity is not down. The lawsuits threaten which sources it can read, so plan for a thinner citation set inside a healthy 200, and detect it in code.

Meta AnalysisAI CodingAPI DesignMonitoringObservabilityTypeScriptDeveloper Tools1 min read · Jul 25, 2026

What Opus 5's Novel-Level Reasoning Benchmark Actually Means

Opus 5 novel reasoning rests on one number, 30.2% on ARC-AGI-3, and it scores action efficiency, not problems solved. Here are the four config lines to change.

Meta AnalysisAI CodingClaude CodeMLComparisonDeveloper Tools1 min read · Jul 25, 2026

WebMCP Has 20× the Interest and Zero Deployments

WebMCP adoption is zero across 111,076 scanned domains while search interest holds near 20 times its January level. Here is what to ship now, and what to watch.

Meta AnalysisMCPWeb DevAPI DesignTrend CommentaryJavaScript1 min read · Jul 25, 2026