Agentic Enterprise
How can bounded AI capabilities become dependable parts of everyday enterprise work? I explore the relationship between workflow, authority, context, and human review.
Current Thinking
Ideas under active exploration.
I keep an active lab around applied AI, organizational cognition, verification, enterprise memory, and new operating models. These are experiments and hypotheses that help me recognize emerging patterns before they become conventional practice.
Verification & Evaluation
What evidence should make us confident enough to act? Experiments connect expected behavior, failure cases, and evaluation to deployment decisions.
Proofloop
An evolving Django prototype for discovery, evidence, hypotheses, and delivery artifacts. The question is how learning from an engagement can improve the next decision.
Agentic BPO
An exploration of service operating models that combine human judgment with agent execution. The focus is bounded work, accountability, and measurable customer outcomes.
Enterprise Memory
Working research in preserving provenance and making experience retrievable. Memorybank explores how useful context can survive tools, sessions, and model changes.
Human + Machine Operating Models
Hypotheses about organizing specialized people and agents around shared outcomes, explicit authority, and feedback.