Member of Technical Staff - Infrastructure

Warp Sciences; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Warp Sciences seeks an infrastructure and systems engineer to build the foundation for production AI workloads. You’ll work across core infrastructure, data systems, and the training and inference stack, ensuring reliability and scalability as AI pipelines scale in California.

You’ll optimize GPU utilization, data throughput, and cost, own production reliability, and collaborate with model teams to turn experiments into dependable production systems.

Qualifications

• Proven track record designing, shipping, and operating production systems at scale.
• Strong fundamentals in distributed systems, Linux, networking, storage, concurrency, and performance engineering.
• Experience building and operating cloud infrastructure, container orchestration, and infrastructure as code.
• Experience with data infrastructure, including distributed processing, pipelines, and storage systems.
• Hands-on experience supporting AI training and inference workloads, including GPU infrastructure and distributed execution.
• Ability to profile systems, identify bottlenecks, and deliver measurable improvements in performance, reliability, or cost.
• AI-native approach: collaborate with model teams and use AI tools in engineering workflows.
• Strong ownership, communication, and judgment when requirements are ambiguous.

Responsibilities

• Design, build, and operate infrastructure for large-scale AI training and inference.
• Build reliable data infrastructure for ingestion, processing, storage, and delivery to training/evaluation workflows.
• Optimize GPU utilization, distributed training performance, inference latency, throughput, and cost.
• Optimize data storage and access for training and inference.
• Own production reliability through observability, capacity planning, incident response, and recovery.
• Improve deployment, orchestration, and developer tooling for rapid progress.
• Collaborate with AI engineers/researchers to turn experiments into dependable production systems.
• Diagnose bottlenecks across compute, networking, storage, data pipelines, and model execution.

Skills

Distributed systems

Linux

Networking

Storage

Performance engineering

Observability

Incident response

Container orchestration

Infrastructure as code

AI workloads

GPU infrastructure

Tools

Cloud infrastructure

Kubernetes

Docker

Terraform

CI/CD pipelines

Monitoring tools

Job description

At Warp Sciences, our mission is to help the world discover new medicines faster—and build the capacity to do it here in America. We bring AI, robotics, and life sciences together in automated labs that design, run, and learn from experiments, with each experiment making the next one better. We’re not developing drugs of our own. We’re building the infrastructure for the people who do.

We’re looking for an infrastructure and systems engineer to build the foundation for production AI workloads. You’ll work across core infrastructure, data systems, and the training and inference stack, making these systems reliable, efficient, and easy to operate as they scale. The inference stack is a combination of local robotics models and perception models and cloud foundation models and the surrounding runtime.

Our strongest requirement is a track record of building and operating production systems that have scaled. You should bring deep systems expertise, practical experience with AI workloads, and the judgment to balance performance, reliability, cost, and engineering speed. You should be ai pilled and using swarm of agents for development.

What you’ll do

• Design, build, and operate infrastructure for large-scale AI training and inference.
• Design & build reliable data infrastructure for ingestion, processing, storage, and delivery to training and evaluation workflows.
• Optimize GPU utilization, distributed training performance, inference latency and throughput, and infrastructure cost.
• Optimize data storage, and access for training and infernece.
• Own production reliability through observability, capacity planning, incident response, and recovery.
• Improve deployment, orchestration, and developer tooling so engineers and researchers can move quickly.
• Partner with AI engineers and researchers to turn experimental workloads into dependable production systems.
• Diagnose bottlenecks across compute, networking, storage, data pipelines, and model execution.

What we’re looking for

• A demonstrated track record of designing, shipping, and operating production systems at scale, with ownership beyond the initial launch.
• Strong fundamentals in distributed systems, Linux, networking, storage, concurrency, and performance engineering.
• Experience building and operating cloud infrastructure, container orchestration, and infrastructure as code.
• Experience with data infrastructure, including distributed processing, pipelines, and storage systems.
• Hands‑on experience supporting and optimizing AI training and inference workloads, including GPU infrastructure and distributed execution.
• The ability to profile systems, identify bottlenecks, and deliver measurable improvements in performance, reliability, or cost.
• An AI‑native approach: comfortable working closely with model teams, understanding how AI workloads behave, and using AI tools effectively in your own engineering workflow.
• Strong ownership, clear communication, and sound judgment when requirements are ambiguous.

Helpful, but not required

• Experience with reinforcement learning infrastructure, rollout generation, or distributed RL training.
• Experience with model serving, batching, caching, quantization, or other inference optimization techniques.
• Experience with GPU scheduling, high‑performance networking, or heterogeneous compute environments.
• Experience with Robotics

What success looks like

You build infrastructure that holds up under real production demand. Training runs use compute efficiently, inference meets latency and throughput targets, data pipelines are dependable, and failures are easy to detect and diagnose. Your work gives the AI team a reliable foundation for shipping and scaling.