My Codex run burned 100,000+ tokens over 8 hours and still handed me an unfinished mess... so I built something that just finishes
Summary
After a Codex run wasted over 100,000 tokens without completing a task, the author built a tool that reliably finishes code generation jobs.
Similar Articles
@RayFernando1337: I finally cracked it! There is a better way to run Codex agents without the slop. As bonus it saves you tokens! Enginee…
Ray Fernando shares a workflow tip for running Codex agents more efficiently, claiming it avoids messy output and saves tokens, and is used by engineers at top FANG companies.
@cjzafir: In the last 17 days, I burned $6k worth of tokens on my Claude 20x and Codex 20x subs. But I got so much work done by j…
The author open-sources a 'lean-thinking' skill for AI coding agents that reduces token usage by up to 35% while maintaining quality, based on personal testing with Claude and Codex models.
@cjzafir: 359M Tokens burned in 72 hours. Cost: $78~ Results: New 240M fine-tuning dataset. Process: > Codex 5.5 as Orchestrator.…
A developer used Codex 5.5 as an orchestrator and Deepseek v4 pro as an executor to generate a 240M token fine-tuning dataset, burning 359M tokens at a cost of only $78.
Codex-maxxing for long-running work
This whitepaper from OpenAI presents practical strategies for using Codex as a persistent workspace to maintain context, manage complex workflows, and sustain progress across long-running projects by breaking goals into verifiable steps.
I measured where my AI coding agents waste tokens, 42% was avoidable. Built a tool to catch it (Claude Code / Cursor / Codex)
The author measured token waste in AI coding agents and found 42% avoidable, then built a tool to catch it. The tool works with Claude Code, Cursor, and Codex.