Machine Learning Infrastructure Engineer

Location: San Jose, CA
Experience: 1–5 years
Team: Applied AI

We’re hiring a Machine Learning Infrastructure Engineer to build the runtime, platform, and operational backbone for modern AI systems. This role is for someone who wants to work on the systems behind the systems: model access layers, routing, serving paths, telemetry, observability, evaluation infrastructure, and the controls needed to make fast-moving AI work reliable in practice.

This is a platform role, but not in the old sense. The work is tightly coupled to how modern AI systems are actually built and used: multiple model providers, agent runtimes, skill and tool layers, inference telemetry, cost-aware routing, AI spend visibility, and governance that is strong enough for real internal adoption.

• Build and improve internal AI infrastructure for LLM applications, agents, retrieval systems, and model-backed engineering workflows.
• Own inference deployment paths across managed and self-serve environments, including access control, monitoring, and operational reliability.
• Build platform layers such as model gateways, routing, runtime integrations, telemetry, and controls for safe execution at scale.
• Develop AI Ops capabilities across evaluation, release readiness, observability, incident triage, regression detection, and cost monitoring.
• Build dashboards, tracing, logging, and alerting for production AI systems, including spend and usage visibility across tools and teams.
• Improve performance and unit economics through routing, caching, batching, failover, and latency/cost optimization.
• Create reusable APIs, SDKs, and platform abstractions that make AI systems easier to deploy, evaluate, govern, and operate.

• 1–5 years of experience in software engineering, ML infrastructure, MLOps, platform engineering, or related backend/infrastructure roles.
• Strong Python plus strong systems instincts.
• Experience with AWS or GCP and real production service ownership.
• Familiarity with inference deployments, model APIs, gateways, serving systems, or runtime infrastructure for LLM/ML workloads.
• Experience with observability, telemetry, reliability engineering, and incident response.
• Understanding of eval systems, release workflows, retrieval-backed systems, and debugging non-deterministic AI behavior.
• Ability to translate messy platform needs into scalable internal infrastructure.

They have built or operated systems where latency, routing, cost, telemetry, and reliability actually matter. They understand that modern AI infrastructure is not just about getting a model endpoint running. It is about building the runtime, visibility, controls, and developer experience that let an applied AI team move fast without losing quality or trust.

The team is building AI-ready infrastructure in the most literal sense: observability, access control, AI spend tracking, secure managed platforms, skill/tool infrastructure, and telemetry that spans requests, tools, models, and outcomes. If you want to work on the platform layer that makes modern agentic systems possible — and do it in a setting where the downstream users are serious engineers with high expectations — this is that role.

The base pay compensation range for this role is between $140,000 - $165,000