Bola Omoniyi

“Built and shipped a production LLM-powered RAG system at Upstart enabling natural-language search across 50k+ scattered internal technical docs. Delivered sub-300ms p95 latency for ~50 active users with strong hallucination safeguards (retrieval-first, thresholds, citations) plus robust testing/monitoring and cost controls (prompt caching cutting API spend ~20%).”

“Backend/infra-focused engineer who owned production systems for distributed ML experimentation (hyperparameter tuning across a cluster with GPU scaling, custom scheduling, and checkpoint-based fault tolerance). Also built and operated a low-latency log validation service using queued async workflows with idempotency, retries/backoff, and strong observability, plus experience building resilient Selenium-based browser automations for complex multi-step web flows.”