Member of Technical Staff — Kernel / Compiler / Communication

<h2 data-start="382" data-end="403"><strong data-start="385" data-end="403">About the Role</strong></h2> <p data-start="405" data-end="553">RadixArk is seeking a<span class="Apple-converted-space">&nbsp;</span><strong data-start="427" data-end="492">Member of Technical Staff — Kernel / Compiler / Communication</strong><span class="Apple-converted-space">&nbsp;</span>to push the limits of performance for frontier AI systems.</p> <p data-start="555" data-end="739">You will work at the lowest layers of the stack — kernels, runtimes, compilers, and communication libraries — to unlock maximum efficiency from modern accelerators and interconnects.</p> <p data-start="741" data-end="956">This role is critical to scaling training and inference across thousands of GPUs, where microseconds and memory bandwidth matter. Your work will directly shape the performance envelope of next-generation AI systems.</p> <p data-start="958" data-end="1109">This is a deeply technical role for engineers who enjoy working close to hardware and solving performance problems that most engineers never encounter.</p> <h2 data-start="1116" data-end="1135"><strong data-start="1119" data-end="1135">Requirements</strong></h2> <ul data-start="1137" data-end="1693"> <li data-start="1137" data-end="1212"> <p data-start="1139" data-end="1212">5+ years of experience in systems, compiler, or performance engineering</p> </li> <li data-start="1213" data-end="1268"> <p data-start="1215" data-end="1268">Strong expertise in CUDA or accelerator programming</p> </li> <li data-start="1269" data-end="1332"> <p data-start="1271" data-end="1332">Deep understanding of GPU architecture and memory hierarchy</p> </li> <li data-start="1333" data-end="1394"> <p data-start="1335" data-end="1394">Experience writing or optimizing high-performance kernels</p> </li> <li data-start="1395" data-end="1459"> <p data-start="1397" data-end="1459">Strong background in compilers, runtimes, or code generation</p> </li> <li data-start="1460" data-end="1539"> <p data-start="1462" data-end="1539">Experience with distributed communication libraries (NCCL, MPI, RCCL, etc.)</p> </li> <li data-start="1540" data-end="1603"> <p data-start="1542" data-end=