Infrastructure / Platform Engineer

Job description

Location: Remote (Australia or New Zealand) Type: Full-time, remote About Neuphonic Neuphonic is building the next generation of voice AI, making natural, real-time speech accessible across devices, languages, and applications. Our technology enables ultra-low-latency text-to-speech, voice cloning, and multilingual speech generation. We're pushing the boundaries of what's possible with speech models, from high-performance cloud inference to running AI directly on everyday devices without expensive GPUs or an internet connection. We're a small, ambitious team of researchers and engineers building technology used by developers and businesses around the world. Our open-source releases have reached millions of downloads, and we're growing quickly. Why we're hiring As our customer base grows, we need to scale our production infrastructure while maintaining the performance, reliability, and low latency that our products depend on. We're looking for an Infrastructure / Platform Engineer to take ownership of our cloud infrastructure and inference systems. You'll work closely with our engineering and ML teams to build scalable, reliable, and cost-efficient systems. A key part of this role is extending our production support coverage beyond UK working hours. We're hiring in Australia or New Zealand to help us work towards 24/7 support for our production customers, so we're looking for someone comfortable taking ownership of live systems, responding to incidents, and making independent technical decisions. Importantly, this isn't just about keeping our systems running. We want someone who will actively look for ways to improve how we build, deploy, and serve our models, continually experimenting, optimising, and evolving our infrastructure as we scale. What you'll own Cloud infrastructure: Build, manage, and scale our production infrastructure across GCP and GPU cloud providers using Kubernetes, Docker, and Terraform. Inference optimisation: Continuously improve inference performance, latency, throughput, GPU utilisation, and infrastructure costs through profiling, experimentation, and iteration, rather than simply maintaining the status quo. Platform evolution: Identify bottlenecks and inefficiencies in how we build, deploy, and serve models, and take ownership of implementing improvements. Production reliability: Monitor critical systems, respond to incidents, and h…