In this two-month update video, Chris confirms that he is still using Claude Code as his primary AI coding tool, despite many new options emerging in the AI coding space. He highlights that Claude Code remains the best tool in his experience, especially due to several new features and workflow improvements that have significantly boosted his productivity. Chris begins by sharing a simple but effective tip: adding the phrase “ultrathink” to prompts for difficult tasks, which instructs Claude Code to think more deeply and often yields better results.
One of the major new features Chris emphasizes is the ability to run background tasks within Claude Code. This allows him to run backend servers directly in the background, giving Claude Code live access to server logs for easier debugging without the need to copy and paste error messages repeatedly. This feature alone has saved him a lot of time and hassle, especially when working with server backends or front-end projects. Additionally, Chris uses Claude Code as a powerful research assistant, leveraging its ability to perform deep web searches and synthesize information from multiple sources to create detailed plans for complex features like PDF extraction.
Chris also discusses the use of MCPs (Modular Cloud Plugins), which enable Claude Code to access external tools and services. He uses MCPs to connect Claude Code to updated documentation and databases, such as Superbase and Firebase, allowing the AI to pull real-time data and even make modifications. This integration has streamlined his debugging process by letting Claude Code directly access user data and identify issues without manual intervention. Furthermore, Chris highlights the GitHub action feature that lets Claude Code review pull requests automatically, catching bugs, security issues, and best practices, which is especially valuable for solo developers without a dedicated code review team.
Another productivity booster Chris mentions is his use of voice dictation through a tool called Whisper Flow. Dictating prompts instead of typing allows him to provide more detailed and natural instructions to Claude Code, improving communication and context. Whisper Flow is particularly adept at understanding developer terminology and even supports tagging files by voice commands. Chris also briefly touches on custom agents within Claude Code, which allow users to define specialized AI assistants for specific tasks, though he finds this feature somewhat overhyped and uses it mainly for iOS development.
Finally, Chris shares a small but handy quality-of-life improvement: custom status lines in Claude Code’s interface. He uses this feature to display the current Git branch and time since the last commit, helping him keep track of his progress at a glance. Overall, Chris is very positive about Claude Code’s rapid evolution and the new features that have enhanced his workflow, especially background tasks, MCP integrations, and voice dictation. He encourages viewers to try Claude Code if they haven’t already and invites them to share their own workflow tips in the comments.
