Staff Machine Learning Engineer, ML Infrastructure
Overview:
As a Staff ML Engineer in our Cloud ML team, you will shape the ML platform that powers real-time computer vision and GenAI workloads at scale. You’ll lead architecture decisions, mentor engineers, and drive reliability and efficiency of production ML systems. You’ll collaborate across teams to raise the bar on deployment, observability, and cost, delivering robust cloud inference that accelerates product features. This role offers impact at scale within a values-driven, collaborative culture.
Compensation / BenefitsFree SimpliSafe system and professional monitoringTotal rewards package (bonus/equity) Medical, retirement, and lifestyle benefitsEmployee Resource Groups (ERGs) Values-driven, inclusive culture
Responsibilities:
Set technical direction for ML infrastructure and drive cloud-native architecture decisions (Kubernetes-based platform with Ray, KServe, Triton, vLLM) Lead design reviews, capacity planning, and reliability for high-stakes ML systemsIdentify and remove deployment bottlenecks (reliability, observability, scaling, cost) Build and operate real-time CV inference pipelines processing live video/events at scaleDevelop and operate LLM/GenAI serving infrastructure and product-feature deployment patternsMentor engineers through reviews, pairing, and guidance; establish best practices for model lifecycle and on-callOwn incident response, postmortems, and platform improvements; define SLOs and observability standardsDocument runbooks and architectural decisions to keep the platform legible and durableCollaborate to raise engineering bar across teams and drive cross-functional alignment
Key requirements:
8+ years of software/ML engineering with production ML systems experienceDeep expertise in Kubernetes-based cloud ML infrastructure; hands-on with Ray for inference; familiarity with KServe, Triton, vLLM, Kubeflow, Argo or similarStrong AWS production experience (EKS, S3, IAM, networking); Kafka, containerized deployments, CI/CD, IaCProven ability to design and operate high-throughput, low-latency inference (GPU-aware scheduling, batching, autoscaling, multi-tenancy) Solid ML fundamentals: training, evaluation, versioning, deployment, monitoring, rollback in productionProficiency in Python; systems language (Go, C++, Rust) a plusStaff-level leadership: cross-functional influence, mentorship, and alignment without formal authorityStrong written and verbal communication; ability to explain tradeoffs to ML scientists, product, and infra teamsStrong written and verbal communicationMentorship and coachingCross-functional collaborationKubernetes (EKS) RayKServe