Software Engineer, ML Serving

📋 Description Architect and implement the TTS serving infra (GPU-based). Optimize models for multi-node/disaggregated serving. Ensure NVIDIA hardware compatibility (Hopper–Blackwell) for on-prem/cloud. Build CI/CD workflows for the serving pipeline. Maintain site reliability: on-call, monitoring, alerts, observability. Provision resources and manage GPU fleet costs. 🎯 Requirements Hands-on with real-time multinode ML serving infra; Dynamo/Triton/vLLM or equivalent. Experience with distributed or disaggregated model serving (Tensor Parallel, Pipeline Parallel, or equivalent). Strong cloud fundamentals: Linux, networking, Docker, Kubernetes. IaC experience with Terraform, Packer, or equivalents. On-call is part of the job; productivity reliability is a shared responsibility. SRE, DevOps, or platform engineering background is a plus. 🎁 Benefits Build the serving infra behind a voice AI company. Direct collaboration with inference, platform, and ML teams. Your work shapes what customers deploy at scale. Meaningful equity upside at an early stage. High ownership, high standards, low bureaucracy. SF / Bay Area presence.