🐋 ContaineRL - Containerize your RL Environments and Agents
How I built ContaineRL, a lightweight toolkit to package RL environments and agents as reproducible, containerized services, exposing them via gRPC with a clean Python API and CLI. It enables modular, language-agnostic RL pipelines, supports scalable deployment, and enforces strong engineering practices (type checking, testing, CI) without sacrificing developer ergonomics.
Nov 1, 2025
🔀 Fast Deep RL on GPU: Adding Multi-Discrete Actions to RSL-RL Library
How I extended the RSL-RL library to support multi-discrete action spaces, enabling new classes of optimization problems beyond continuous control. This post shares the motivation, implementation details, and benchmark results compared to Stable Baselines 3.
Aug 1, 2025
🛡️ Reinforcement Learning for Decision-Level Interception Prioritization in Drone Swarm Defense
Using Reinforcement Learning to coordinate multiple kinetic effectors and defend a sensitive area from large-scale kamikaze drone swarms, trained entirely in a custom-built, high-fidelity simulation.
Jul 1, 2025
⚔️ DIAMBRA: Building a Tournament Platform for DeepRL Agents
A platform for training, submitting, and evaluating DeepRL agents in arcade-style environments with real-time tournaments, leaderboards, and community-driven competition.
Jun 1, 2025