Infrastructure Engineer

Kotoba's speech models serve 2,000–3,000 new app users a day and are licensed to Fortune 50 companies. We're hiring an Infrastructure Engineer to own the cloud infrastructure underneath all of it. Location: San Francisco or Tokyo. Our footprint is expanding from Japan into the United States, and the central challenge of this role is designing and operating infrastructure across regions and continents. Everything is managed as code with Terraform and built cloud-agnostic. This isn't a narrowly scoped infrastructure role — you'll take ownership of the service end to end. ■ About Kotoba Kotoba is a generative AI company on a mission to become the default for voice AI in East Asia. At our core is a low-latency, high-accuracy speech translation model that connects conversations so naturally it feels as though both speakers share the same language, supporting Japanese, English, Korean, Chinese, Spanish, and other major language pairs. We also build ultra-low-latency speech-to-text and text-to-speech models that run everywhere from the data center to edge devices, and we license this foundational technology to Fortune 50 companies and major US tech firms. We work from two hubs: Tokyo and San Francisco. Our own product, the Kotoba app, is available on iOS and Android. Since launch it has grown to a steady 2,000–3,000 new downloads per day and reached No. 1 in its App Store and Google Play category, ahead of the likes of Google Translate. Enterprise adoption is accelerating in Japan, and the app has supported nearly 100 live events including SusHi Tech Tokyo. Kotoba was founded in 2023 by two Japanese generative AI researchers with PhDs from top US universities. We've raised over ¥3 billion (roughly US$23M) from prominent VCs in Japan and the US — including Kindred Ventures and Globis Capital Partners — and from the corporate venture arms of leading US and Japanese enterprises. We also receive strong government support in Japan for AI model training. ■ What you'll do Maintain, manage, and design Kotoba's cloud infrastructure, managed as code with TerraformDesign and build cross-regional infrastructure spanning Japan and the United StatesManage infrastructure across both the application and the API/SDK platformDevelop a thorough understanding of our applications and contribute to end-to-end testing and troubleshooting, taking responsibility for the overall service rather than infrastructure aloneDeploy customer-specific AI models and manage testing endpointsManage real-time communications infrastructure using WebSocket, WebRTC, and gRPC, meeting the quality and performance demands of real-time voiceOptimize GPU and CPU resource configurations and manage infrastructure costsBuild and monitor the infrastructure required to maintain certifications such as ISO, and oversee security and internal controls including VPN infrastructure and access management ■ What we're looking for Required Experience designing infrastructure across multiple continents and regions — this is the hardest technical problem in the role, and we welcome candidates who want to take it onExperience designing and operating cloud infrastructure, particularly on Azure and AWSHands-on experience with Kubernetes and HelmKnowledge of real-time communication technologies such as WebSocket, WebRTC, and gRPC — infrastructure experience limited to HTTP-based systems won't be enough hereThe instinct to look beyond infrastructure in isolation: understanding our applications and taking ownership through end-to-end testing. We're looking for a versatile engineer who keeps a holistic view of the system rather than working to a narrowly defined specialtySubstantial hands-on experience and the ability to own the company's cloud infrastructure as a wholeBusiness-level English. Native-level fluency isn't required, but English is essential for daily communication with the wider team and with colleagues running PoC projects Preferred Knowledge of infrastructure security, including experience supporting certifications such as ISOExperience operating infrastructure for AI model–based servicesExperience managing and optimizing GPU resourcesInfrastructure experience with high-traffic, API-based systemsExperience implementing and managing logging systems for service monitoring and operations ■ Location San Francisco or Tokyo.