RadixArk jobs

A Head of Business Development role at RadixArk, an AI infrastructure company in Palo Alto. The position involves building and leading a BD team, managing customer relationships at scale, identifying partnership opportunities, and synthesizing market intelligence. Requires 10+ years of BD/sales/partnerships experience and building teams from scratch. Explicitly requires on-site work 5 days/week in Palo Alto office—not remote.

RadixArk is hiring a Performance Engineer in Palo Alto, CA to optimize LLM inference and training systems for production scale. You'll work on the performance-critical path of SGLang, Miles, and RadixArk's infrastructure, focusing on latency, throughput, GPU utilization, memory efficiency, and cost-per-token across real customer workloads on NVIDIA, AMD, Google TPU, and cloud platforms. Key responsibilities include benchmarking LLM workloads, debugging performance regressions, optimizing batching and scheduling, building internal profiling tooling, and translating customer needs into concrete tuning strategies. You need a strong systems engineering background with GPU or distributed systems experience, proficiency in Python and C++, familiarity with profiling tools, and ideally hands-on experience with CUDA, kernel optimization, or LLM inference concepts like KV cache and speculative decoding. RadixArk is an infrastructure-first startup founded by engineers who've shipped production AI systems and created open-source projects like SGLang, backed by infrastructure investors and partnered with NVIDIA, Google, and AWS. The role requires comfort debugging messy real-world performance issues across software and hardware layers—this is deep systems work, not generic benchmarking.