Dev runs Gemini coding all day without hitting Pro limits; agent nails sound design
lipeng0820 · x · 2026-11-21
Blogger lipeng0820 shares a day of full-stack development with Gemini (referred to as Gemini Neon): a Pro account survived heavy frontend + backend demands without hitting rate limits.
Highlight: when a notification chime needed replacing, the agent located playRoseCompleteChime() in DotMatrixAudioEngine.ts, diagnosed why it sounded flat (two plain sine waves, 659Hz→880Hz, no overtones, volume coefficient 0.14), then produced an in-chat A–F listening panel of six replacement sounds, all synthesized in pure Web Audio — zero licensing risk, zero load delay.
Screenshots also show the usage panel: Gemini weekly limit 89% left, 5h limit 62% left, while Claude/GPT models' 5h limit was fully exhausted, and an unbilled "Gemini 4 Argon" weekly limit sitting at 100% — suggesting per-model quota separation.
More from coding & agent
- Teaching a retired pharmacist Claude: model tiers, prompt templates, and the 40-prompt rot — rubenhassid · 2026-11-21
- Pro-tip: pair your agent with a second agent whose only job is to challenge it — craigbalding · 2026-11-21
- AI gateways are becoming the key control layer in production AI stacks: 8 capabilities — goyalshaliniuk · 2026-11-21
- Mnemos 2.0: agent memory designed to preserve what matters for the next session — RileyRalmuto · 2026-11-21
- InferenceBench ranks agents on LLM serving speedup: Claude Opus 5.5 tops at 12.08x, agents lose to simple hyperparameter search — maksym_andr · 2026-11-21
- 4 architecture patterns to tame multi-agent system reliability, from a 27-year reliability engineer — bibryam · 2026-11-21