Even Opus 5.5 hasn't internalized the bitter lesson when writing LLM systems
brandon_galang · x · 2026-10-20
The quoted post observes that LLMs themselves haven't internalized the bitter lesson: when asked to build LLM-powered systems, they keep encoding rigid knowledge in deterministic code instead of knowing where to switch to prompts. The poster jokes that even Opus 5.5 doesn't realize how good it is—so relax about impostor syndrome.
More from coding & agent
- ClaudeBot crawler observed regularly fetching sitemap files for URL discovery — gaganghotra_ · 2026-10-20
- Evals are the definition of done: 'Demo-ready' isn't 'monetize-ready' for AI agents — leveragedupside · 2026-10-20
- Cloudflare Workers often break subdomain-as-subfolder redirects for VPN users, causing 404s — gaganghotra_ · 2026-10-20
- Codex's new suggested-next-prompt is 'scary good', devs say — charliermarsh · 2026-10-20
- Stripe Atlas adds MCP support so AI agents can incorporate your startup — jeff_weinstein · 2026-10-20
- Cross-Agent Verification: Use Codex as a Read-Only Reviewer for Claude's Code — umeshonai · 2026-10-20