100% of the code was written locally on a 16GB RTX GPU using gemma-4-12b-qat, a model…
This is a dev post classified by AGK Intelligence as AI dev tools (a tutorial), kept by the Dev Radar because it carries real work, not commentary.
100% of the code was written locally on a 16GB RTX GPU using gemma-4-12b-qat, a model with a roughly 7 GB footprint. Across 12 human-guided turns, I continued refining the application until exhausting its 64K-token context window.
Posted by David Hendrickson (12.8k followers) 50 min ago · 2 likes · 314 views · view the original post on X. Kept by the Dev Radar as AI dev tools.
More dev work like this
- 写长篇故事，别再靠一堆 Prompt 硬撑了。 — @bkdgiffug
- People still don't get how much Hermes does. — @HermesWatcher
- Missing an agent tool? Nodecode can write it and keep it. — @DanKornas
- I’ve moved my local LLM testing from manually coordinating/executing model tests in LM… — @PaulGug
- 同时开几个 Claude Code 干活，最怕它们改到同一份代码上。用 Git 自带的工作树隔开又麻烦，新建一个就得把分支名敲三遍。 — @GitHub_Daily
- 🔥最新！OpenCode Browse 正式上线！AI Agent 控制你的真实浏览器，含登录状态、插件、标签页！ — @NFT_Chen
- 为什么我们不在手机上塞一个 SSH 终端来管 Coding Agent？ — @mylifcc
- This is me building a fully native iOS I made on my phone from start to finish (not sped… — @anshnanda
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 32k posts from 5k X accounts over the last 21 days, 9.6k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-07 06:42 UTC. Full method.