Mathematical benchmark exposing the massive performance gap between real agents and LLM wrappers. Rigorous multi-dimensional evaluation with statistical validation (95% CI, Cohen's h) and reproducible methodology. Separates architectural theater from real systems through stress testing, network resilience, and failure analysis.

Breath-based AI agent architecture

A Recursive Ontological Framework for Cognitive Design, Neurodivergence Modeling, AI Co-Development, and "Structural AI"

Embodied agent that learns from human teaching; hybrid agent architecture with computer vision, AI planning, knowledge graph reasoning & generalization , inverse kinematics. Supported by the DARPA GAILA program.

Case Study of the laboratory plant myJoghurt - implemented with JaCaMo BDI agents embedding an engine for processing the NPL(s) language and a set of capabilities for approaching the sanctioning norm enforcement process

First domain agent in A2A architecture. Family life management with JSON-RPC 2.0 agent-to-agent protocol.

A lightweight, pluggable memory backend for agent-based simulations. Supports temporal data, experience replay, and persistent state logging

An open-source engineering blueprint for defining and designing the core capabilities, boundaries, and ethics of any AI agent.

🏥 HealthGuard AI: Multi-agent clinical monitoring system reducing hospital false alarms by 85% using Gemini 2.0 Flash. Built for Google AI Agents Intensive Bootcamp 2025. Features 5 specialized agents, custom MCP medical tools, real-time sepsis/cardiac detection, and Docker deployment.

AutoMCP: Dynamic Tool Synthesis via Code Pattern Mining and Self-Regenerative Agent Architectures

A collection of the messy problems I've hit, the weird fixes I found, and the refactors that finally made things behave. Basically, my debugging diary, but useful.

Hierarchical agent orchestration system with JSON-RPC 2.0 unified architecture

A Python implementation of the canonical agent architecture: a while loop with tools. Build production ready AI agents with purpose-built tools, comprehensive tracing, and async patterns.

Patent-pending EVM v1.1 — dual-layer emotional vector memory architecture for AI, enabling stable personality evolution, resonance fields, and long-term coherence. Developer Edition (CC BY-NC-ND 4.0).

Michael Lorenz, M.Sc. Enegery & Environment Technology – AI Engineer, Cognitive Technologist & Digital Strategist

🤖 Build reliable AI agents in Python using the canonical agent architecture for a clean and debuggable coding foundation.

Multi-Agent Kubernetes Cost Optimization System

🤖 Benchmark AI agent capabilities, bridging the gap between hype and reality with clear metrics and insights for informed development decisions.