Inference Optimization Engineer

Hedra is seeking an Inference Optimization Engineer to collaborate with its research team on enhancing the speed and efficiency of state-of-the-art visual models during inference. The role involves profiling model architectures, identifying performance bottlenecks, and implementing optimizations across various layers of the technology stack, including algorithms, runtimes, and GPU kernels. This position requires deep technical expertise in ML systems and GPU computing to drive measurable improvements in latency, throughput, and memory efficiency.