Design and operate end-to-end infrastructure for intracortical neural data, from ingestion and validation through preprocessing, versioned dataset releases, and training data loaders. Preserve scientific metadata and provenance while implementing signal processing, quality checks, storage optimization, parallel processing, and monitoring. Collaborate with neuroscience, machine learning, infrastructure, and IT teams to support scalable, reliable model training and sensitive research-data management.

Own and optimize deep learning training performance across GPUs, cloud infrastructure, and distributed environments. Profile workloads, develop custom GPU kernels, tune PyTorch and training systems, improve memory and compute efficiency, and design scalable distributed training strategies. Collaborate with researchers and infrastructure teams to preserve numerical correctness while improving speed, cost, reliability, and scalability. Build benchmarking, checkpointing, recovery, and regression-testing workflows for long-running AI/ML experiments.

Leads architecture, roadmap, execution, and team development for high-performance BCI software spanning embedded, edge, and distributed systems. Establishes engineering practices, testing, infrastructure, and documentation for regulated medical devices; coordinates hardware, clinical, AI, cybersecurity, and regulatory dependencies; manages hiring, resources, budgets, and technical delivery while remaining engaged in architecture and critical problem-solving.