Remote job
Engineering Architect, Platform & Ecosystem
Job details
About this role
Role overview A remote, full-time architect role focused on turning new RISC-V silicon into a working software platform. You'll lead Linux bring-up, SDK delivery, and ecosystem enablement for AI inference and HPC workloads, bridging kernel-level work with developer-facing tooling that ships alongside the chip.
Responsibilities - Lead platform bring-up on new RISC-V silicon, including Linux kernel, bootloader, device tree, and driver enablement. - Build and ship SDKs for silicon partners: toolchains, AI and HPC libraries, documentation, and examples. - Enable and optimize DPDK, networking stacks, MPI, and system libraries for RISC-V targets. - Integrate AI frameworks (PyTorch, ONNX Runtime) and HPC stacks (Fortran runtime, numerical libraries, Julia packages) into cohesive platform deliverables. - Drive RVA profile compliance and conformance testing against ecosystem standards. - Engage with open-source communities to build relationships and influence upstream projects.
Requirements - Deep experience bringing up Linux on new hardware platforms. - Strong knowledge of DPDK, networking stacks, MPI, and HPC system libraries. - Production experience building SDKs that serve both AI and HPC developer audiences. - Familiarity with AI frameworks such as PyTorch and ONNX Runtime, and HPC stacks including Fortran runtimes, numerical libraries, and Julia packages. - Understanding of RISC-V RVA profiles and platform specifications. - Ability to context-switch across layers (kernel, SDK packaging, DPDK optimization, Julia integration) while maintaining depth. - Active engagement with system-level open-source projects.
Nice to have - DPDK contributor history or networking-stack development experience. - Experience integrating modern language ecosystems (Julia, Rust) alongside traditional stacks (Fortran, C/C++).
Benefits and work setup - Remote, full-time position with the opportunity to build a complete platform stack on freshly fabricated silicon for workloads spanning AI inference to HPC simulation.