Teams at Runware

Own the end-to-end network architecture for GPU data center fabrics, backbone and edge delivery, cloud and hybrid connectivity, operations, automation, capacity planning, and vendor relationships. Design high-performance Ethernet and InfiniBand networks, optimize latency and resilience, establish observability and incident practices, and lead through technical authority. The role also sets standards, mentors engineers, shapes hiring, and uses AI tooling to accelerate network design, automation, analysis, and documentation.

Lead global infrastructure operations for Runware’s GPU pod fleet across the US and Europe. Own site deployment, Linux bare-metal systems, storage, networking, reliability, capacity, security, vendors, budgets, and incident response. Build and lead a distributed infrastructure team while establishing automated, repeatable deployment and operating practices. Partner with hardware and platform engineering to improve operability, and use AI-driven automation to scale a high-performance infrastructure organization.

Own the end-to-end storage architecture for a distributed AI supercomputer, including object storage, high-throughput pod tiers, NVMe caching, replication, and model distribution. Build and benchmark systems, optimize data movement from storage to GPU memory, establish reliability and security practices, and automate pod bring-up and operations. Partner with inference, platform, hardware, vendors, and site teams while supporting capacity planning, monitoring, recovery, data residency, and storage on-call.