Self Review vs Peer Review: Dev Trashes His AI-Built Game HUD as "Very Very Slop"
Dimillian · x · 2026-10-23
Indie developer Dimillian shares a screenshot contrasting self review with peer review: first a glowing AI-style self-assessment of a new game HUD (winged skull crests on elite plates, boss crowns, exact health bars, a fixed layout bug), followed by his own brutal human review: "This is insanely bad... the rank badges are sooo bad. Very very slop."
The post nails the gap between AI-generated output's self-evaluation and real human review, with "slop" as the cutting verdict.
More from coding & agent
- Vibecoding your own LLM inference engine: GGUF in WebGL2 ends in debug hell — CodeMichaelD · 2026-10-23
- Dev builds an agent-human 'software factory' against full-AI delivery — Pavan_Belagatti · 2026-10-23
- Open-source Harness Engineering repo turns coding-agent guidance into mechanical guardrails for Claude Code — blaizedsouza · 2026-10-23
- Harness Engineering: How OpenAI and LangChain Got Agents to Ship a Million Lines of Code — blaizedsouza · 2026-10-23
- Agent Relay, a YC-backed open-source tool, lets your AI agents talk to each other directly — ycombinator · 2026-10-23
- Java beats Rust in head-to-head word processor benchmark: 5.8x faster typing, smaller binaries — rseroter · 2026-10-23