Inference Performance Engineer

About us We're an inference cloud built for AI ASICs, generating tokens 5–7× faster than existing GPU infrastructure at a fraction of the price. We closed our oversubscribed seed round with $97M in compute allocation and $200M in hardware financing underway. We believe that when something is important enough, no obstacle is insurmountable. If you thrive on extreme ownership, outsized impact, and relentless optimism, this is the place for you. About the role Build the inference runtime that powers our ASIC cloud including batching, KV cache optimization, scheduling, and APIs. You'll be working with experts from top manufacturers operating at the frontier of AI hardware design. What you'll do Build and improve the inference runtime that serves our ASIC hardware Own scheduling, continuous batching, KV cache optimization, prefill, and decode separation Optimize tokens/sec, TTFT, p99 latency, and cost per token Collaborate with hardware and compiler teams to update kernels and operators Maintain the OpenAI-compatible API surface Benchmarking and regression testing What you'll need BS in Computer Science or related field 3+ years of software engineering experience: Rust, Go, or Python Solid fundamentals in concurrency, memory, and tail latency Familiarity with modern LLM inference: transformers, attention, KV cache, batching, speculative decoding, quantization Experience with model serving: vLLM, TGI, SGLang, TensorRT-LLM, llama.cpp, or custom runtimes What we'd like CUDA, ROCm, Triton, kernel-level work, or experience with non-NVIDIA accelerators Built and scaled an OpenAI-compatible API in production What we offer Competitive cash compensation Generous stock options 100% paid medical, dental, and vision insurance for employees Flexible PTO Paid Holidays Equal Employment Opportunity We're an Equal Opportunity Employer and do not discriminate on the basis of any protected status under applicable law.