Apple is where individual imaginations gather together, committing to the values that lead to great work. Every new product we build, service we create, or Apple Store experience we deliver is the result of us making each other’s ideas stronger. That happens because every one of us shares a belief that we can make something wonderful and share it with the world, changing lives for the better. It’s the diversity of our people and their thinking that inspires the innovation that runs through everything we do. When we bring everybody in, we can do the best work of our lives. Here, you’ll do more than join something — you’ll add something!
As a Senior Staff/Principal Engineer on the Foundation Model Compute Infrastructure team, you will architect, build, and operate large-scale infrastructure powering foundation model training, fine-tuning, evaluation, and inference.
You will tackle challenging problems in distributed scheduling, accelerator utilization, cluster management, fault tolerance, and large-scale job orchestration. You will work with cloud and hardware partners to bring new TPU generations into production and build the tools and services that enable researchers to launch, monitor, debug, and profile their workloads without needing deep infrastructure expertise.
You'll develop model inference and fine-tuning services, build secure, scalable sandbox infrastructure for running AI agents, code execution, and reinforcement learning environments at massive scale, and work closely with foundation model researchers and engineers to improve performance, reliability, scalability, and developer productivity.
This is a deeply hands-on technical leadership role. You will write, review, and debug production code, lead complex architectural decisions, and use modern AI-assisted development tools to accelerate engineering execution and raise the quality bar across the team.
Lead by example through hands-on coding, debugging, and solving the team's most challenging architecture and implementation problems.
Design and build distributed scheduling, quota, and capacity management systems to maximize accelerator utilization across diverse workloads and teams.
Improve fault tolerance, reliability, and recovery for large-scale training and inference workloads spanning thousands of accelerators, including automated hardware failure detection and remediation.
Architect and operate Kubernetes-based infrastructure, including accelerator provisioning, node lifecycle management, cluster orchestration, and autoscaling.
Collaborate with cloud and hardware partners to qualify, integrate, and deploy new generations of TPU hardware.
Design, build, and operate secure, high-density sandbox infrastructure for executing untrusted, model-generated code and AI agent workloads, with strong isolation, fast startup, and efficient resource utilization to support reinforcement learning (RL) training, agentic evaluation, and tool use.
Build self-service developer tools, APIs, and observability systems for job submission, monitoring, debugging, and performance profiling.
Design and develop scalable services for foundation model fine-tuning and inference.
Partner with researchers, ML framework engineers, and security teams to translate evolving research requirements into robust production infrastructure.
Mentor engineers, drive technical design reviews, and establish engineering and operational best practices across the team.
8+ years of experience designing and building distributed systems or large-scale infrastructure, with demonstrated technical leadership.
Strong hands-on experience with Kubernetes, cloud infrastructure, and operating production systems at scale.
Proficiency in Python, Go, C++, or other programming languages used for infrastructure and distributed systems.
Experience with accelerator infrastructure and machine learning training or inference workloads.
Excellent written and verbal communication skills, with experience authoring technical design documents and driving collaboration across teams.
Hands-on experience operating TPU, GPU, or other large-scale accelerator fleets.
Experience scaling distributed training or inference workloads across thousands of accelerators.
Familiarity with JAX, XLA, and distributed machine learning frameworks, including performance analysis and profiling.
Experience developing LLM-based agents or AI-assisted developer and infrastructure operations tools.
Deep knowledge of distributed scheduling, resource allocation, multi-tenant quota management, or cluster autoscaling.
Job description provided by Apple.
View the original posting
