AI-Powered Development: Tenex Engineer Unveils Custom Tools for Streamlining Claude Code Workflows

The integration of AI into software development is rapidly transforming the landscape for engineers. CJ Hess, a software engineer at Tenex, is at the forefront of this revolution, building custom tools that enhance the capabilities of AI models like Claude Code. In a recent episode, Hess delved into his workflows and the tools he's developed to optimize his AI-assisted development process.

One of the key tools Hess discussed is Flowy, a custom visual planning tool designed to overcome the limitations of traditional ASCII diagrams. Flowy converts JSON files into interactive visual mockups and flowcharts, providing a more intuitive and efficient way to plan complex UI and animation workflows. This approach offers cognitive benefits compared to text-based planning, allowing developers to visualize and manipulate designs more effectively. Flowy draws inspiration from tools like Mermaid diagrams, Figma, and Excalidraw, but is tailored specifically for integration with Claude Code workflows.

Model-to-Model Code Review

Hess also shared his innovative approach to code quality assurance using model-to-model comparison. By leveraging both Claude Code and Codex (referred to as 'Carl' in the podcast), he can compare the code generated by each AI model, identifying potential errors or areas for improvement. This method helps ensure that the final code is robust and reliable, taking advantage of the strengths of different AI models. Bypassing permissions in controlled environments to speed up development is also discussed.

The podcast also explores the broader evolution of developer environments with AI, emphasizing the value of building a custom ecosystem of tools around Claude Code. Hess highlights the importance of extending the development environment with custom skills and workflows to maximize efficiency and productivity. He also mentions the potential of Google's Project Genie and other emerging AI technologies to further enhance the development process.

Hess's work with Flowy and model-to-model comparison showcases the transformative potential of AI in software development. By creating custom tools that augment the capabilities of AI models, developers can streamline workflows, improve code quality, and ultimately build more innovative and impactful applications. The resources mentioned in the podcast, including Claude Code, Claude Opus 4.5, Cursor, Obsidian, and GPT-5.2 Codex, provide a valuable starting point for developers looking to explore the possibilities of AI-assisted development.

Companies like Orkes and Rovo are also mentioned, providing enterprise platforms for reliable applications and AI agents that can adapt to the specific needs of a business.

For those interested in learning more about CJ Hess's work, you can connect with him on LinkedIn and X (formerly Twitter), or explore his projects like ChatPRD. He also shares insights into his company, Clairevo, on LinkedIn and X.