Sr. Inference Optimization Engineer (local / edge runtime)

This role involves performance engineering for AI inference engines targeting edge and local computing environments—PCs, on-premises systems, and resource-constrained hardware rather than cloud datacenters. You'll profile and optimize inference pipelines across different hardware configurations, working with open-source engines to reduce latency and memory overhead while maintaining model quality. The work bridges the gap between cutting-edge AI capabilities and practical deployment on hardware people actually own, making hybrid AI products economically viable. You'll collaborate with post-training teams on quantization strategies, contribute patches upstream, and establish performance baselines across hardware tiers. This role suits engineers with strong systems-level optimization experience who are curious about inference internals and excited about making AI more accessible and efficient at the edge.