DeepSeek engineer relections on RSI - burying my talent to yesterday

A DeepSeek engineer published a blog post reflecting on recursive self-improvement following the release of DeepSeek v4.1, which the post says raised the ability of small models to a new level. The engineer wrote that AI is improving much faster than anyone expected, contrasting it with the first ChatGPT, which could only chat simply with a few thousand tokens. The post was translated from the original blog link.

Note - This is translated from the actual blog link right at the bottom. A few days ago, DeepSeek v4.1 was released. It raised the ability of small models to a new level. AI is improving much faster than anyone expected. From the first ChatGPT that could only chat simply with a few thousand tokens o