AgentsSeptember 14, 2026via InfoQ AI/ML
Presentation: Decision Models in Agentic Architectures: From Production to Agent Skills
Why it matters
A production blueprint for building accountable agentic systems where business logic stays auditable and governance stays tight — solving the non-determinism problem that blocks agent deployment in regulated industries.
Key signals
- Gap: non-deterministic output and lack of accountability in high-stakes agentic decisions
- Solution: integrating DMN (Decision Model and Notation) with LLMs, agent skills, and NeMo guardrails
- Outcome: auditable, deterministic agentic architectures with business logic ownership and architectural governance
- Use case context: enterprise/regulated environments requiring decision traceability
- Integration of DMN (Decision Model and Notation) with LLMs and agent skills
- Focus on auditable, deterministic outputs in high-stakes decisions
- NeMo guardrails for architectural governance
- Separates business logic (owned by leaders) from engineering architecture (owned by engineers)
- Addresses non-determinism and accountability gap in enterprise AI deployment
The hook
Enterprise agents need deterministic decisions. DMN models + LLMs create the audit trail.
Alex Porcelli discusses the critical gap in enterprise AI: non-deterministic output and lack of accountability in high-stakes decisions. He shares how integrating DMN decision models with LLMs, agent skills, and NeMo guardrails creates auditable, deterministic agentic architectures - allowing busine…