Jobs at ElastixAI

Design, develop, and maintain core ML inference platform components including model deployment, optimization pipelines, and benchmarking/simulation workflows. Collaborate with systems and cloud engineers, and build APIs/tools to ensure scalable, reliable, and hardware-efficient inference solutions.

Design and implement IR transformations, graph optimizations, kernel lowering, and code generation for novel hardware. Decompose LLM/transformer workloads into primitives, build performance models and profiling tools, collaborate with ML and hardware teams, prototype end-to-end improvements from framework passes to custom kernels, and shape system architecture for an inference engine.

Design, operate, and evolve ElastixAI's Kubernetes and multi-cloud inference infrastructure. Run accelerated ML workloads at scale, build deployment and automation tooling, harden AWS/GCP/on-prem systems, partner with ML/runtime teams to productionize models, optimize costs and reliability, and participate in on-call rotation.

Design and optimize a low-level AI inference serving stack: customize open-source frameworks, build model partitioning/scheduling, integrate with proprietary accelerators, profile and optimize across Python orchestration to C++ kernels and drivers, and enable PyTorch-native deployment tooling.

Design and implement a novel AI inference engine through hardware-software co-design. Collaborate with ML, software, and cloud engineers; model PPA trade-offs; contribute to RTL design, simulation, verification, and roadmap planning to optimize inference for modern LLM workloads.

Conduct and lead applied ML research to develop efficient generative models and AI inference techniques. Prototype and implement novel methods, run experiments, write production-quality code, analyze results, and collaborate with engineering teams to transition research into scalable production systems.