Skip to content
CoRISE

From Tacit Knowledge to AI Workflows: Observe Decisions and Exceptions

Make criteria, exceptions and responsibility inspectable before translating human judgment into prompts.

1 min read
  • workflow
  • ai
  • data
Open table of contents

Conclusion

Make criteria, exceptions and responsibility inspectable before translating human judgment into prompts.

Context

Routine expert judgment often contains undocumented exceptions and interpretation. Observe actions and decisions, then separate them into inspectable units.

Design and verification scope

Assess the following responsibilities and boundaries when designing and verifying a configuration.

  • Observation
  • Explicit decision criteria
  • Workflow
  • Data
  • Systems and AI
  • Human review

Decision rationale

Compare practitioner explanations with observed work and identify when apparently similar inputs lead to different decisions. Record criteria, exceptions, information needs and responsible reviewers. Compare rules, search, conventional systems and AI after separating deterministic work from uncertain judgment.

Trade-offs

Explicit criteria require observation and review and cannot capture all tacit knowledge. Too few exceptions misrepresent practice; too many become unmaintainable. Preserve escalation conditions and assign responsibility for updating criteria as work changes.

Limitations

Review observations, decision examples and counterexamples, practitioner feedback and approvals. Interviews alone do not formalize a whole workflow.

These cases provide attributed design context. They do not establish that the proposed experiments or configurations were delivered in those engagements.

Contact

Tell us about your engineering challenge.

Talk with CoRISE about the design, implementation and operation of your systems.

Start a Conversation