Arnaldo Sepulveda

Lead Engineer, Keystone Applied Intelligence

Enterprise AI engineer. I build governed retrieval and agent systems for regulated industries: on-premises deployment, evidence-backed answers, query-time role-based access control, and tamper-evident audit trails.

Before Keystone, I spent 12+ years at Genesys delivering enterprise contact center platforms for regulated and public sector customers. MScE in Electrical Engineering from the University of New Brunswick, with a thesis applying machine learning to smart grid load control.

Current work

Keystone AI: a governed RAG platform. Public technical project and live demo at getkeystone.ai. Currently extending the system to governed agentic actions: tool authorization by role, action audit trails, human-in-the-loop approval gates.

Eval baseline (KDAT-001B, 2026-04-11): P@1 0.75, MRR 0.79, adversarial ACL 8/8 blocked, fail-closed 5/6.

Find me