Debanjan Basu
Where theoretical physics meets practical AI
Berlin, Germany
I'm a Senior Python Developer working at the intersection of AI systems, distributed computing, and backend infrastructure. My path here wasn't linear—it wound through five years of theoretical physics research at TU Clausthal, data science at an IoT startup, and eventually into building production AI systems.
What I carry from physics isn't just mathematical fluency. It's a particular way of seeing: finding the minimal description that captures a system's essential behavior, recognizing when a problem has hidden symmetries, knowing when to zoom in on details and when to step back and look for universal patterns.
These days, I build AI agent systems, web scraping infrastructure, and observability pipelines. I'm fascinated by the emerging discipline of making AI systems that are not just capable, but observable, debuggable, and composable.
Series
Production Django Task Queue
Building a ~300 LOC task queue on Django ORM from prototype to production — memory leaks, fork pitfalls, pessimistic locking, and security hardening.
Terminal Power User
Kitty terminal, kittens, shell integration, Starship prompt, and turning the terminal into a complete development environment.
Wayland Desktop Mastery
Niri's scrolling paradigm, unified shortcuts across compositors, DankMaterialShell, and building a cohesive Wayland desktop.
Berlin's Transit Crisis
How reunification, austerity, broken funding, and low wages created BVG's crisis — and what proven European models show about the way out.
A Deep History of Bengali Culture
আদি বাঙালি ইত্যাদি — A personal journey through the deep history of the Bengali language: the people, migrations, and forgotten civilizations folded into the words we use every day.
Standalone Posts
Two Claude Code Power Features You Should Be Using
Custom status lines for ambient awareness and git worktrees for parallel AI-assisted development sessions.
SSH in 33 Seconds? Optimizing for India-to-Bulgaria VPN Connections
Diagnosing and fixing SSH latency over high-latency VPN links, with focus on the TCP-over-TCP problem and connection multiplexing.
What I Write About
The Practical Side
Development environment deep-dives, agent harness architectures, LLM orchestration patterns, and observability for AI systems.
The Theoretical Side
Physics-inspired approaches to understanding neural networks—phase transitions, renormalization group theory, information geometry.
Other Wanderings
Ancient history, travels to places where the past feels tangible, and whatever else I find worth thinking through in writing.
Get in Touch
I'm always interested in conversations about AI systems, physics-informed machine learning, or elegant solutions to hard problems.