Rafael Cisneros

Machine Learning Engineer @Quantum Ventura Inc

San Francisco, CA, US

Shipped CNRTurbo, a streaming threat-detection stack processing 15 million events/day; cut alert latency from 7s to under 0.5s and container size by 20%.• Built an online threat-detection ensemble combining NEAT-based neural architecture search (PyTorch + CUDA) with neuromorphic Akida SNN models; boosted F1 by 3% on multiclass intrusion logs while reducing inference cost on edge nodes.• Designed end-to-end pipeline: Zeek → Kafka → Flink → PyTorch service (Docker, FastAPI, MLflow, Django) → React/Next.js dashboard via WebSockets.• Containerized and deployed models on proprietary servers; implemented Prometheus/Grafana monitoring, achieving 99.9% uptime and drift detection in under 6 min.• Partnered with a six-person DevOps and analyst team, cutting mean-time-to-detect incidents by 42%.

Tecnun - Universidad de Navarra

Bachelor of Engineering - BE, Telecommunication Engineering and Computer Science

IESE Business School

Master in Management, Business Management

University of San Francisco

Master of Science - MS, Data Science