My Agentic Coding Setup, July 2026 (domenic.me)

In a recent exploration of AI-assisted development techniques, a developer unveiled a robust coding setup that leverages advanced tools and models to streamline their workflow. By transitioning from a corporate environment, they have created a system that allows them to fix production bugs remotely, even from their phone, using a Linux VM driven by generative models like Claude and ChatGPT. Key components of this setup include using Tailscale to create a secure private network for seamless SSH access among devices, and configuring a Linux VM to maximize the potential of these AI tools. This new approach is significant for the AI/ML community as it showcases the innovative integration of AI in real-world coding environments, emphasizing autonomy and efficiency. By minimizing manual approvals and enhancing agent capabilities, the developer can maintain parallel workstreams without bottlenecks, illustrating the potential for AI to significantly enhance productivity in software development. The setup is a time capsule of current tech practices, as it incorporates features like lightweight client applications for agent management, dynamic development servers, and Git workflows that adapt to collaborative coding, highlighting how the interplay between AI, coding environments, and network configurations is shaping the future of software development.