Member of Technical Staff — Training

<h2 data-start="322" data-end="343"><strong data-start="325" data-end="343">About the Role</strong></h2> <p data-start="345" data-end="471">RadixArk is seeking a<span class="Apple-converted-space">&nbsp;</span><strong data-start="367" data-end="407">Member of Technical Staff — Training</strong><span class="Apple-converted-space">&nbsp;</span>to build and scale the systems that train frontier AI models.</p> <p data-start="473" data-end="734">You will work on large-scale distributed training infrastructure for LLMs and generative models, pushing the limits of scale, efficiency, accuracy and reliability across 10k, or 100k+ of GPUs. This role sits at the intersection of ML, systems, and performance engineering.</p> <p data-start="736" data-end="820">Your work will directly impact how next-generation AI models are trained and scaled.</p> <p data-start="822" data-end="938">This is a deeply technical, high-impact role for engineers who enjoy solving hard systems problems at extreme scale.</p> <h2 data-start="945" data-end="964"><strong data-start="948" data-end="964">Requirements</strong></h2> <ul data-start="966" data-end="1597"> <li data-start="966" data-end="1067"> <p data-start="968" data-end="1067">3+ years of experience in ML systems, or large-scale training infrastructure</p> </li> <li data-start="966" data-end="1067"> <p data-start="968" data-end="1067">Experience building or operating large-scale agentic post-training systems.</p> </li> <li>Experience working on training / inference correctness or other precision-related problem</li> <li data-start="1310" data-end="1390"> <p data-start="1312" data-end="1390">Experience debugging performance and stability issues in large post-training jobs</p> </li> <li>Experience improving training or inference efficiency.</li> </ul> <h3 data-start="1604" data-end="1623"><strong data-start="1608" data-end="1623">Strong Plus</strong></h3> <ul data-start="1625" data-end="2061"> <li data-start="1625" data-end="1679"> <p data-start="1627" data-end="1679">Experience training 100+ billion-parameter models</p> </li> <li>Experience with train / inference optimization for large-scale RL or other production workload.</li> <li data-start="1680" data-end="1756"