CoregentisAI

Working on open standards for reliable multi-agent systems.

I build Coregentis, an open research & engineering effort focused on: - protocol design for agent lifecycles - deterministic execution models - state / plan / trace governance - reproducibility & safety for LLM-driven systems

Creator of MPLP (Multi-Agent Lifecycle Protocol), an Apache-licensed specification for agent interoperability and lifecycle governance.

Previously spent years building large-scale distributed systems, automation pipelines, and developer tooling for high-availability environments.

Interested in: - agent runtimes & orchestration - schema-first system design - verification, observability & invariants - reproducible AI engineering - open protocols → ecosystem infrastructure

Here to learn, debate, and exchange ideas with people building the future of agentic computing.