Real-Time Cloud Infra Engineer for Robotics

Weave Robotics; San Francisco (CA); On-site; USD 150,000 - 190,000; Full time

Job summary

Weave Robotics in San Francisco is seeking a Cloud Infrastructure Engineer to own the networking stack that connects robots to GPUs, teleoperators, and customers’ companion apps. You’ll design protocols for reliable home networks and the backend that manages a growing fleet across multiple customers.

The role targets real-time, low-latency challenges—from routing to GPUs to keep inference fast, to stable teleoperation streams and secure cloud services that scale with demand.

Qualifications

• 3+ years building backend or media infrastructure in production, with real ownership of systems under load.
• Real-time networking depth: built/operated low-latency streaming systems over UDP, NAT traversal, congestion control, and protocol literacy beyond SDKs.
• Cloud infrastructure knowledge: experience operating production cloud infrastructure using IaC and containerization tools such as Terraform/OpenTofu, Ansible, Docker, and Kubernetes, with hands-on production deployment, monitoring, and debugging.
• Networking fluency: deep understanding of UDP/TCP, NAT traversal, STUN/TURN/ICE, congestion control, packet loss, jitter, and home networking realities.
• Cloud & device security: familiarity with TLS/mTLS, PKI, certificates, KMS/HSMs, secrets management, device identity, and credential provisioning and rotation for connected devices or distributed systems.
• Production data experience: schema design, migrations, and operational care for databases customers depend on.

Responsibilities

• Build distributed infra: Design backend services that coordinate robots, cloud compute, teleoperators, and customer applications across unreliable networks and partial failures. Build for fault tolerance, idempotency, consistency, scalability, and graceful degradation as the fleet grows.
• Serve cloud inference: Build low-latency serving infrastructure for models robots call mid-task, including connection management, GPU scheduling and utilization, autoscaling, load balancing, and graceful degradation when connectivity or compute capacity is constrained.
• Build the product backend: Develop APIs and services for fleet, customer, and robot state, backed by well-designed data models and production databases. Own schema evolution, migrations, performance, backup, and recovery as these systems scale.
• Secure robot-to-cloud communication: Build authentication, authorization, device identity, credential provisioning and rotation, and secure communication between robots and cloud services, partnering with Security from architecture through deployment.
• Build the real-time transport stack: Own low-latency media and control transport end-to-end: connection lifecycle, NAT traversal on real home networks, encrypted transport, congestion control, jitter handling, and the latency budget teleoperation depends on.
• Fleet telemetry. Ingestion, storage, and query of logs and metrics from every robot in the field, serving engineering, fleet operations, and ML data needs.
• Own production reliability: Define SLOs and build monitoring, alerting, deployment, and incident-response systems that keep critical fleet services available and make failures diagnosable.

Skills

Backend engineering

Real-time networking

Cloud infrastructure

Networking fundamentals

Device security

Production databases

Tools

Terraform/OpenTofu

Ansible

Docker

Kubernetes

Job description

Weave Robotics in San Francisco is seeking a Cloud Infrastructure Engineer to own the networking stack that connects robots to GPUs, teleoperators, and customers’ companion apps. You’ll design protocols for reliable home networks and the backend that manages a growing fleet across multiple customers.

The role targets real-time, low-latency challenges—from routing to GPUs to keep inference fast, to stable teleoperation streams and secure cloud services that scale with demand.