I used AI to reconstruct more than two decades of professional work scattered across old publications, broken links, archives, indexes, and surviving records.

The difficult part was not finding possible matches; it was preventing AI from turning incomplete evidence into false certainty. I built a workflow that treats archive reconstruction as a verification problem rather than a search problem.

The process separates verified work from probable matches, resolves same-name conflicts, deduplicates reprints and syndicated copies, preserves source provenance, and treats missing years as research gaps rather than permission to invent records.

The result was a source-tracked professional archive containing 250+ recovered records across more than 20 years. The workflow can be adapted to journalism, research, consulting, creative work, speaking histories, or any long-term body of professional output.

I use the following research-archivist workflow:

• Never invent titles, dates, publications, URLs, or authorship.
• Separate records into VERIFIED, PROBABLE, and REJECTED.
• Treat same-name individuals as separate identities until evidence proves otherwise.
• Deduplicate syndicated copies, reprints, mirrors, indexes, and cached pages.
• Preserve provenance for every verified record.
• Mark gaps rather than filling them.
• Distinguish “not found,” “unverified,” and “confirmed absent.”
• State uncertainty explicitly.

Step-by-step

1. Ask for names and bylines, employers, publications, career dates, subject areas, known titles, source files, URLs, and any same-name conflicts.
2. Build an identity fingerprint.
3. Establish verified seed records.
4. Search iteratively across multiple query patterns and sources.
5. Classify every result.
6. Resolve identity conflicts.
7. Deduplicate records.
8. Normalize metadata.
9. Build a chronology.
10. Identify and search gaps.
11. Run a provenance and error audit.

The final output includes:

• Verified Master Archive
• Probable / Needs Verification
• Rejected / Other Person
• Gap List
• Source Notes
• Final Audit

The archive is never described as complete unless the evidence supports that conclusion.