Does anyone else find the way they are writing this full marketing hubris that definitely misconstrues how most people would interpret this. Codex isn't a "model" that is self improving, it's using GPT to write code that is in a wrapper program that also uses GPT. Sure it's kind of neat loop for development, but why are they anthropomorphizing it so much? People designing chips don't say that computers are self evolving, even anthropic just says that claude (the model) writes most of claude code. Heck, you could use claude or any llm to write code for codex..
The "GPT in a wrapper" framing is exactly right.. If any sufficiently capable model can be the engine, the value isn't the model, it's the orchestration and context management around it.
Codex is a product bet that OpenAI can bundle faster than others can plug in competing models.
The more I think about this the more it just seems like a marketing piece.
"Everyone is running a bunch of agents all at once... writing code, and doing code and security review" (paraphrasing
Okay...? What's the role back rate on code pushed to staging or production? How many false positives do they get a day from the reviews? Or worse, completely missed issues. I run security scans in CI, what value add does an agent provide?
"New teammates are trained in the morning and expected to push to prod in the afternoon" (again paraphrasing)
This is either totally insane or pure marketing BS. No way you have a brand new teammember pushing to prod the same day they start. The only way this is even feasible is if the DevOps team has an incredible roll back story, which would be a far more interesting article.
What's the opinion of actual developers? The view of a bunch of managers is completely useless.
Does anyone else find the way they are writing this full marketing hubris that definitely misconstrues how most people would interpret this. Codex isn't a "model" that is self improving, it's using GPT to write code that is in a wrapper program that also uses GPT. Sure it's kind of neat loop for development, but why are they anthropomorphizing it so much? People designing chips don't say that computers are self evolving, even anthropic just says that claude (the model) writes most of claude code. Heck, you could use claude or any llm to write code for codex..
It would be interesting to see a breakdown of how much this usage would cost a normal dev team paying for GPT.
The "GPT in a wrapper" framing is exactly right.. If any sufficiently capable model can be the engine, the value isn't the model, it's the orchestration and context management around it.
Codex is a product bet that OpenAI can bundle faster than others can plug in competing models.
The more I think about this the more it just seems like a marketing piece.
"Everyone is running a bunch of agents all at once... writing code, and doing code and security review" (paraphrasing
Okay...? What's the role back rate on code pushed to staging or production? How many false positives do they get a day from the reviews? Or worse, completely missed issues. I run security scans in CI, what value add does an agent provide?
"New teammates are trained in the morning and expected to push to prod in the afternoon" (again paraphrasing)
This is either totally insane or pure marketing BS. No way you have a brand new teammember pushing to prod the same day they start. The only way this is even feasible is if the DevOps team has an incredible roll back story, which would be a far more interesting article.
What's the opinion of actual developers? The view of a bunch of managers is completely useless.