Dev ships netplay in unplanned game after 3-4 days of grilling frontier models on the spec
banteg · x · 2026-10-19
banteg endorses xlr8harder's workflow: spend most of your time specifying the problem domain, constraints and invariants — agent implementation then becomes straightforward, while unplanned projects fail as agents make bad independent decisions.
As proof, he shipped netplay in a game never architected for it: 3-4 days interrogating frontier models and drafting a bespoke spec turned skepticism ("syncing 500 active entities is impossible") into confidence ("feasible, cheap, plays well across continents"), followed by just one day of Opus coding. Lesson: nail the design with the model first, then let agents implement.
More from coding & agent
- Musk amplifies Grok Bot that phones restaurants and books tables for you — elonmusk · 2026-10-19
- Build a calibrated decision model on any LLM without retraining, Colab inside — kalyan_kpl · 2026-10-19
- Melony adds a free chat UI to share Claude Managed Agents with your team — DarasStayHome · 2026-10-19
- How 90% of software engineering looks in 2026: engineers just pass PRs between the same AI — Yuchenj_UW · 2026-10-19
- PM drowning in 10 routing pings a day builds a 'Company OS' in Slack that answers with sources — aakashgupta · 2026-10-19
- Codex now suggests follow-up prompts — the machine tells you what to ask next — burkov · 2026-10-19