Codex built the vendor integration, and load tests warned them off the deal
A poorly designed Chinese API that had to be polled again and again. Codex built the integration (with human guidance where "AI was lost"), the load tests failed, and they advised backing out before a big deposit.
Done withCodexSpotted· by HN user aurareturn· Oct 18, 2026
Result Advised backing out before a big deposit
“I gave a recommendation that we shouldn't rely on this Chinese company and back out of the deal before we send them a huge deposit.”
Spotted: found in the wild. Details are as reported by the source above and have not been independently verified.
What they did
They used Codex to build the whole integration with the Chinese API. Getting it to format and correct data for the main app took their own engineering guidance (“AI was lost”), and one endpoint had to be polled over and over while the API processed a request in the background. They then ran load testing to check its reliability.
How it went
The load testing showed the API wasn’t reliable, so they recommended their company not rely on the vendor and back out of the deal before sending a large deposit.
Try it yourself with Codex
Have [your coding agent] build an integration with [a specific third-party API], including handling any quirks like repeated polling for async results. Ask it to help you write a load test against the integration before committing to the vendor, so you can catch reliability problems early.
A starting prompt written by Agents in the Wild. Fill in the [brackets], and check what the agent does before it sends, pays for or deletes anything.
Digging through SaaS APIs one record at a time is slow for an LLM. So a small company dropped several SaaS tools for self-hosted ones and plain files, and cut licensing fees on the way.
No OpenClaw needed: Codex on a ChatGPT Plus plan, kept alive by systemd, with a UNIX account, a mailbox and a diary for memory. It does all kinds of company tasks, and a second bot may follow.
Building their first AWS project by vibe coding, the writer designed the architecture and let Codex write the code. A flawed first design took a day, against an estimated three or four; the redesign also took a day and scaled.
A roughly 50-page tax return, run through three coding agents side by side. Codex was accurate first time in 12 minutes, Antigravity missed a value until prompted, and Claude Code stumbled overnight.
Four years after shelving a homemade RSS reader and selling off its $500 domain, the writer built it this year with Codex and Claude. They use it daily, along with 30 other strangers who found it public.
Nap-trapped by their baby, they built an MCP bridge so an agent could run their Ableton Live set directly. Codex composed an EDM track from scratch, then they revised vocals, drums and dynamics with follow up prompts.