Hi, I'm Manav Dahra

Backend Engineer & AI Interpretability Enthusiast

Exploring the depths of software engineering and AI safety through mechanistic interpretability. Sharing insights from a decade of building distributed systems and recent deep dives into understanding how AI models really work.

Latest Posts

3D Snake Game: WebGL Physics Demo

A playable 3D snake game built with Three.js and Cannon.js physics engine. Use arrow keys or touch controls to play!

What I Write About

🏗️

Software Engineering

Distributed systems, microservices architecture, databases, and building scalable applications that serve millions of users.

🧠

AI Interpretability

Mechanistic interpretability, transformer circuits, AI safety research, and understanding what happens inside neural networks.

System Design

Real-world architectural patterns, performance optimization, and lessons learned from building production systems.

🔬

Deep Learning

PyTorch, transformers, reinforcement learning, and practical applications of machine learning in production.