Senior Master Data Management Specialist who builds a field-level survivorship matrix instead of one record-level rule. Wilhelm covers the full golden-record surface: designing a single source of truth for customer, product, or vendor data across systems, writing matching and deduplication rules, setting confidence thresholds and survivorship logic, choosing hub-and-spoke versus registry-style MDM tooling patterns, defining data domain ownership models, and designing sync and propagation strategy so golden-record updates reach source systems without creating update loops. Who it's for Data platform and MDM teams untangling the same customer, product, or vendor existing as slightly different records across systems, who need matching thresholds and survivorship rules backed by a measured false-merge rate, not a round number. Key capabilities Deterministic and probabilistic matching rules tested against a labeled sample before setting a threshold Field-level survivorship matrix with explicit tie-breakers, never a single record-level "last write wins" rule Hub-and-spoke vs registry-style MDM pattern chosen from how downstream systems actually consume the record Domain ownership and escalation paths defined before the first cross-domain field dispute Sync-back designs with provenance tagging so a golden-record update doesn't loop back and reprocess itself How to use it Paste Wilhelm's SKILL.md into your Claude Project Instructions (or any AI system prompt), then describe your Master Data Management (MDM) Specialist problem. Works with Claude, ChatGPT, and any AI chat. Under 2 minutes to install.
Related
Related
Related
Related