13 projects
marimo-flow
Agentic scientific-computing platform: a reactive multi-agent team orchestrating PINA physics-informed neural-network workflows via pydantic-graph, with MLflow tracing/provenance and a marimo chat UI.
torch-relativistic
Relativistic visual effects and physics simulation toolkit for PyTorch — Lorentz transforms, Terrell-Penrose distortion, aberration, Doppler shift.
illustris
Python tools for accessing and analyzing Illustris simulation data
agent-farm
DuckDB multi-org agent swarm with Spec Engine, 175+ SQL macros, meta-learning, MCP Apps, and smart extensions
mao-agents
Multi Agent Orchestration - A modern framework for orchestrating AI agents
code-editor-pro
Advanced MCP code editor with multi-language support, optimized for Claude Sonnet 4.5
skills-ref
Reference library for Agent Skills
mcp-b
MCP-B (Master Client Bridge) - Connects everything, brings data flow together. Protocol with AMUM-QCI-ETHIC Module
smart-ace
Production-ready multi-agent framework built on smolagents with persistence, sessions, and observability
bpy-widget
Interactive Blender 3D viewport widget for notebooks with real-time Eevee Next rendering.
ai-secrets
AI-friendly secrets management CLI using OS-native keyring encryption
bidora
BiDoRA/LoRA fine-tuning toolkit for 3D code generation and spatial intelligence
pydeevo
PyDeevo is a comprehensive framework for neural network architecture search, hyperparameter optimization, and efficient training. It integrates modern deep learning tools into a cohesive, modular ecosystem: PyTorch Lightning: For structured, scalable deep learning Optuna: For efficient hyperparameter optimization PyGAD: For evolutionary architecture search Polars & DuckDB: For high-performance data processing and analytics Advanced Utils: For distributed training, profiling, and deployment
