Openai is rolling ‘Codex’ for Chatgpt, an AI agent that automatically and represent the programming tasks for software engineers.
Openai is clearly not claiming that Codex will eventually replace junior software engineers.
Instead, the company states that Codex can help developers to achieve more by handing over their tasks to various agents.
The idea is to move faster with development and become more productive with AI, but how does Codex work?
According to Openai, Codex is based on Codex-1, a new version of Chatgpt based on the current O3 model, but has been adapted for coding, resulting in an increase in accuracy.
Codex draws codebase from Github and reflects the existing PR style closely.
This can write a new code, propose a bridge requests, and can run each task in its sandbox.
“The task is usually between 1 and 30 minutes in completion, depending on the complexity, and you can monitor the progression of Codex in real time,” OpenIAI noted In a blog post.
“Once the Codex completes a task, it completes its changes in its environment. Codex provides verification of its functions through the terminal log and test output quotes, allowing you to detect each step taken during the completion of the work.”
Codex is rolling out, but only when you take a pro subscription, which costs $ 200 per month.
Based on the analysis of 14M malicious tasks, search for the top 10 MITERAT & CK techniques behind the 93% attacks and how to defend them against them.
