Anthropic: A 400000 session study shows that non programmers have a success rate similar to software engineers when using Claude Code Odaily Planet Daily News: Anthropic's latest report on large-scale intelligent agent programming shows that with the assistance of Claude Code, the importance of programming skills themselves has decreased, and management and domain knowledge have become key factors affecting the results. Claude Code users are active for an average of 20 hours per week. Among the top ten major professions, the success rate of non programmers using intelligent agents to write code has narrowed to less than 7% compared to professional software engineers, with management positions having a slightly higher success rate than professional programmers. Users dominate about 70% of planning direction decisions, while intelligent agents undertake about 80% of specific execution decisions. Expert level conversations can trigger 12 actions and output 3200 words in a single session, which is more than 5 times that of novice level conversations. The dropout rate for novice sessions when encountering errors is 19%, while the dropout rate for intermediate and expert users when encountering problems is 5% to 7%. Within 7 months, the proportion of debugging sessions used to repair damaged code decreased from 33% to 19%, and the proportion of direct deployment, data analysis, and writing non code documentation doubled. The average estimated value of tasks increased by about 25%.