Teams at quadric.io

Build cycle-level Python performance models of AI inference workloads for a novel GPNPU. Model tensor placement, tiling, memory hierarchy, sharding, and communication. Calibrate models against simulators and traces, analyze trade-offs (latency vs throughput), and produce technical studies that inform architecture and product decisions.

The AI Inference Engineer will port, optimize, and benchmark AI models for Quadric's platform while providing support and documentation.

The AI Kernel Engineer will develop and optimize AI kernels for Quadric's platform, enhancing performance across various hardware configurations and providing technical support.