Software Development Manager - Compiler (AWS)

Amazon; Cupertino (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Amazon Annapurna Labs seeks a Software Engineering Manager to lead the Deep Learning Compiler Team in Cupertino, Seattle, or Austin. You will direct senior compiler engineers, drive optimization algorithms, and partner with AWS ML services to deploy on CPU, NPU, GPU, and custom AI hardware.

You will shape feature roadmaps, promote best practices in code generation and graph optimization, and help bring new compiler technologies to market within AWS ML initiatives.

Qualifications

• 5+ years of engineering team management experience.
• 9+ years of direct engineering experience.
• 4+ years of designing or architecting scalable systems and reliability solutions.
• Experience partnering with product or program management.
• Deep understanding of compilers, including resource management, instruction scheduling, code generation, and compute-graph optimization.
• Strong software design fundamentals and system-level coding skills with emphasis on graph theory and performance techniques.

Responsibilities

• Lead, mentor, and grow a team of compiler engineers.
• Own the design and implementation of compiler optimization algorithms for neural network workloads on CPU, NPU, GPU, and custom AI hardware.
• Collaborate with hardware and software partners to integrate new instruction sets and scheduling strategies.
• Engage with product and program management to shape feature roadmaps and define scope.
• Champion best practices in resource management, scheduling, code generation, and graph optimization.
• Drive high-quality, maintainable code through rigorous design reviews and testing.
• Represent the team at cross-functional meetings and contribute to the broader AWS ML ecosystem.

Skills

Engineering management

Compiler design

Performance optimization

Graph theory

System-level coding

Resource management

Education

PhD or MS in CS/CE

Tools

XLA

TVM

MLIR

LLVM

Job description

The Product: AWS Machine Learning accelerators lead AWS innovation. The Inferentia chip delivers best‑in‑class inference performance at the lowest cost, while Trainium provides the highest training performance with teraflops of compute power for ML in the cloud. This is enabled by the AWS Neuron SDK, which includes an ML compiler, runtime, and native integration into popular frameworks such as PyTorch and JAX. AWS Neuron and Trainium are used at scale by customers and partners across many industries.

Team Overview

The Amazon Annapurna Labs team builds silicon and software solutions for AWS customers. The team spans silicon engineering, hardware design, verification, software, and operations, all focused on cloud scale and rapid innovation. Members work in small, agile groups and gain experience across multiple disciplines.

Role Overview

We are seeking a Software Engineering Manager to lead the Deep Learning Compiler Team. As a Manager III, you will direct a group of experienced compiler engineers developing optimization algorithms and deploying a new compiler that targets AWS custom hardware. You will collaborate closely with AWS ML services teams, participate in pre‑silicon design, and help bring new products and features to market.

Responsibilities

• Lead, mentor, and grow a team of compiler engineers.
• Own the design and implementation of compiler optimization algorithms for neural network workloads on CPU, NPU, GPU, and custom AI hardware.
• Collaborate with hardware and software partners to integrate new instruction sets and scheduling strategies.
• Engage with product and program management to shape feature roadmaps and define scope.
• Champion best practices in resource management, scheduling, code generation, and graph optimization.
• Drive high‑quality, maintainable code through rigorous design reviews and testing.
• Represent the team at cross‑functional meetings and contribute to the broader AWS ML ecosystem.

Basic Qualifications

• 5+ years of engineering team management experience.
• 9+ years of direct engineering experience.
• 4+ years of designing or architecting scalable systems and reliability solutions.
• Experience partnering with product or program management.
• Deep understanding of compilers, including resource management, instruction scheduling, code generation, and compute‑graph optimization.
• Strong software design fundamentals and system‑level coding skills with emphasis on graph theory and performance techniques.

Preferred Qualifications

• PhD or MS in Computer Science, Computer Engineering, or related field.
• Experience troubleshooting/debugging AI hardware or deploying large language models on GPUs, Neuron, TPU, or similar accelerators.
• Familiarity with XLA, TVM, MLIR, LLVM, and deep‑learning framework design.
• Active participation in open‑source communities, either as a leader or code contributor.

Location & Compensation

The role is available in Cupertino (preferred), Seattle, or Austin. Compensation includes a base salary range, sign‑on payments, and restricted stock units, adjusted for experience, qualifications, and location.

EEO Statement

Amazon is an equal‑opportunity employer and does not discriminate on the basis of protected veteran status, disability, or other legally protected status. Same applies to all applicants, including those from Los Angeles County.