Claude Code Product Leader Cat Wu Leaves Anthropic

Cat Wu, who played a leading role in developing Anthropic’s Claude Code product, has left the company after approximately two years. Wu announced her departure on November 3, 2026, revealing that her final working day had been October 30. In her farewell message, she thanked the teams behind Claude Code, Cowork and Tag, as well as the users whose feedback helped shape the products. She described building Claude Code into an industry-defining product as the journey of a lifetime. Wu did not disclose the reason for leaving or identify her next professional destination. Her departure marks a notable change for Anthropic’s product organisation, particularly because she was closely involved in turning Claude Code from an internal experiment into a widely used AI coding assistant.

Wu joined Anthropic in 2024 and became one of the central figures responsible for Claude Code’s product development alongside its creator, Boris Cherny. Claude Code was initially introduced as a research preview in February 2025 and gradually developed into a powerful AI coding tool capable of writing, modifying, testing and running software. Its growing popularity has helped establish Anthropic as a major competitor in the AI-assisted software development market. Wu also worked on Cowork, Anthropic’s product designed to extend similar agent-based capabilities beyond programming into general workplace tasks. Her work involved translating developer feedback and internal research into practical features that could make AI systems more useful in everyday professional environments.

In her departure announcement, Wu reflected on how dramatically the use of Claude Code had changed since its early development. She explained that users initially relied heavily on manually approving permission requests and mainly used the tool to answer questions or write software tests. Today, she said, users increasingly trust Claude’s judgement to approve routine actions automatically, allowing it to perform large software migrations and build new products. This transition illustrates the broader movement towards autonomous AI coding systems that can handle complex, multi-step engineering tasks with less direct supervision. However, greater autonomy also creates important questions about security, permissions and human oversight, particularly when AI agents are allowed to modify important software systems.

Wu’s departure follows other exits involving members of the Claude Code team. Thariq Shihipar and Anthony Morris also announced their departures from Anthropic in October 2026. Shihipar had worked on Claude Code for more than a year, while Morris was associated with the development of its desktop experience. Neither publicly provided a detailed explanation connecting their decisions to internal disagreements or organisational problems. Anthropic’s leadership has also not publicly established that the departures are related. Nevertheless, the exits have attracted attention because they involve people who contributed to one of the company’s most commercially important products. The timing is particularly significant as Anthropic prepares for a reported major initial public offering, placing greater attention on its leadership, product strategy and ability to retain experienced employees.

Despite the personnel changes, Anthropic continues to expand its AI development tools as competition with OpenAI and other technology companies intensifies. Claude Code faces growing competition from OpenAI’s Codex, GitHub Copilot and other AI-powered programming platforms. The company is also expanding Cowork and developing systems that allow multiple AI agents to work on different tasks. Wu’s departure therefore comes at an important moment for Anthropic, as it attempts to maintain the momentum behind its coding products while preparing for its next stage of growth. Although her exit raises questions about future product leadership, there is currently no confirmed evidence that it will disrupt Claude Code’s development. The company’s ability to maintain its engineering talent, improve its products and sustain developer confidence will remain important as the AI coding market becomes increasingly competitive.