Show HN: Zest – A hands-on simulator for Staff+ system design scenarios (staff-engineering-simulator-880284904082.us-west1.run.app)

Zest, a newly launched hands-on simulator, allows users to validate their skills in designing and managing complex AI infrastructures, particularly in Distributed Systems. By simulating staff-level system design scenarios, the platform covers critical areas such as control plane internals, scheduling mechanisms, and container networking interfaces. Users can dive into distributed training techniques, including optimizations through NCCL and GPU kernel enhancements, which are vital for scaling AI workloads efficiently. This simulator is significant for the AI/ML community as it bridges the gap between theoretical knowledge and practical application, enabling professionals to sharpen their expertise in dealing with real-world infrastructure challenges. Key technical elements, such as KVCache utilization, continuous batching, model sharding, and advanced networking practices like Envoy filter chains, eBPF, and mTLS, are incorporated to provide realistic scenarios. Powered by Google's Gemini 3 Pro, Zest presents a robust tool for engineers to enhance their capabilities in designing scalable, efficient AI systems.