Former ByteDance engineer: Chinese AI practitioners generally rely on US agent tools, domestic coding agents are "completely impractical."

robot
Abstract generation in progress
AIMPACT News, April 24 (UTC+8). According to Beating monitoring, Zhang Chi said in an interview that he uses Claude Code and GitHub Copilot’s Agent mode in his day-to-day work. His students use Claude Code for training monitoring: it automatically checks GPU status, and when an error is found, it automatically debugs and restarts; it performs a review every 30 minutes. “Then we can go to sleep.” When asked whether all AI practitioners in China are using Claude Code, he first said, “Not all.” He said he is more accustomed to using Copilot, but then confirmed that AI practitioners in China generally use U.S. Agent tools. He said bluntly that the coding Agents of Chinese models are “completely impractical,” based on his personal experience rather than a systematic evaluation. ByteDance’s overseas teams can use Cursor directly, and from this he speculated that ByteDance’s future large models may be partially built with the assistance of U.S. Agent tools. Zhang Chi also discussed a personal dilemma: Claude Code is so useful to him that he is considering whether he “still needs to train doctoral students,” but he worries that if he doesn’t train the next generation of researchers, there will be no one left to do research in the future. (Source: BlockBeats)
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pinned