UltronChat.ai
DM automation and an AI agent for Instagram and Messenger, in production.
About
Agentic Infrastructure Architect
I build the infrastructure that lets AI run companies.
Bruno Bracaioli is the creator of AISAC and of the Governed Agentic Workloop concept, and the author of the Workloop Spec. His work focuses on Agentic Infrastructure Architecture: the discipline of designing systems where AIs can operate with autonomy without taking authority over their own limits.

Thesis
AI systems can hold autonomy inside their own loops. Interactions between loops, and any action with an external effect, must respect the policies, rules and deterministic gates the organisation defines.
Path
Three movements, in the order they happened. Each one left something that is still running.
01
B2Tech started as a software house: client products, architecture, security and CI/CD from the first commit. That is where the standard comes from — what survives in production, not what demos well.
02
Then came systems that run on their own, with real users and a bill to pay: conversation automation, niche products, and a fleet of MCP servers wiring agents into marketing, analytics, content and cloud operations.
03
Running a company with agents exposed the real problem: it is not a better model that is missing, it is infrastructure and governance. Out of that came the Governed Agentic Workloop, the Workloop Spec, AISAC and AI Brain — the memory that travels across AIs.
Authorship
A system of two or more autonomous agentic loops interacting through explicit channels under a common authority, with deterministic gates on every interaction and externally effecting action.
Publicly formalised by the Workloop Spec — Agentic Workloop Specification, version 0.1.0-rc5 (2026-08-18), CC BY 4.0, DOI 10.5281/zenodo.22004648.
Read the Workloop SpecEcosystem
The system that creates and operates AI systems: intent turned into governed, versioned and continuously operable systems.
Execution
DM automation and an AI agent for Instagram and Messenger, in production.
A virtual pet powered by your GitHub activity.
A persistent memory layer shared across AIs through MCP, keeping knowledge and context available across sessions, agents and compatible clients.
Model Context Protocol servers that wire AI agents into real marketing, analytics, content, cloud and business operations — hosted, with writes guarded by preview and confirmation.
Validation
Postgraduate student at PUC — Software Architecture, Data Science and Cybersecurity (completing 12/2026).
Public profiles