Riccardo Bronzuoli
Independent AI Systems Builder
I design and prototype AI systems focused on autonomous agents, cognitive architectures, memory, reasoning, evaluation and developmental AI.
Experimental autonomous AI agent architecture integrating:
- custom decoder-only Transformer
- memory systems
- reasoning and planning
- world modelling
- agent runtime
- tool execution
- permission controls
- OS sandboxing
- automated evaluation
The 2.7M-parameter Transformer is intentionally a small proof-of-concept / smoke-test model used to validate the surrounding architecture rather than compete with frontier LLMs.
→ [View project]
Research and architectural proposal exploring how an artificial agent could progressively develop representations, memory, self-models, social models, preferences and autonomy from an initially complete but immature functional architecture.
This project currently exists as documentation and research design; it has not yet been implemented.
→ [View project]
AI Agents · Cognitive Architectures · Memory · Reasoning · Planning · Evaluation · AI Safety · Developmental AI
Currently studying Nursing in Italy while independently developing AI systems and research projects.
Portfolio · CV · LinkedIn