back to work
case study · 2025
Leaderboard
Data Intelligence MVP
- year
- 2025
- role
- solo
- duration
- MVP
- status
- archived
- stack
- 6 tech
- modules
- —
The Brief
Most leadership-effectiveness signals are buried in fragmented public data — filings, scraped pages, scattered APIs that don't share IDs. Leaderboard's MVP fuses them into a single queryable knowledge graph ready for LLM analysis, so stakeholders can uncover organizational risks that no individual source surfaces alone.
The work behind the demo is the unification engine: programmatic entity reconciliation across sources with no shared schema, enrichment pipelines that normalize messy filings, and an indexed graph base that makes every downstream query cheap. Engagement is private. Sole engineer.
Stack
PythonTypeScriptNode.jsPostgreSQLAWS LambdaAWS S3