Systems Architect — I design the infrastructure that runs businesses.
I build data pipelines, scenario engines, and automation systems. The kind of backend infrastructure that replaces manual processes, eliminates operational hours, and scales without babysitting.
Multi-Agent RAG Platform — Production-grade retrieval-augmented generation with knowledge graph enrichment, multi-model routing, and cost-controlled tool-calling agents. 447 tests, 95% coverage. Built with Python, FastAPI, pgvector, Neo4j, and Redis.
Transaction Reconciliation Engine — Automated financial transaction matching across payment gateways and bank feeds. 4-rule scoring cascade with fuzzy matching, processes 4,000 transactions in under 5 seconds. Built with Go, PostgreSQL, and Redis.
CourtVision — NBA player prop predictor. Simulates performance shifts when teammates sit using usage vacuum modeling. Built with Python, SQLite, and REST APIs.
NFLQB — NFL quarterback performance analysis pipeline. Custom composite metrics, correlation analysis, and automated PDF reporting. Built with Python.
Full case studies, architectural breakdowns, and engineering deep-dives at kingsleyonoh.com.



