Full-Time

AI Compiler Engineering Lead

Distinguished Engineer

Posted on 9/3/2024

Ampere

Ampere

1,001-5,000 employees

Manufactures high-performance processors for cloud computing

Data & Analytics
Enterprise Software

Compensation Overview

$195.4k - $325.6kAnnually

+ Bonus + Equity Awards

Senior, Expert

Santa Clara, CA, USA + 1 more

More locations: Portland, OR, USA

Hybrid role requiring in-office presence.

Category
Backend Engineering
Embedded Engineering
Software Engineering
Required Skills
Python
Tensorflow
Git
Linux/Unix
Requirements
  • Master's or PhD degree in Computer Science, Electrical Engineering, Mathematics, or a similar quantitative field and 15 years of overall software engineering experience.
  • Strong Proficiency in Python and C/C++ Languages: Bring your expertise and passion for programming with at least 4 years of hands-on experience in Python and C/C++, driving innovative solutions through powerful code.
  • Experience with Large-Scale Software Systems: Showcase your background working on substantial software projects, especially in the realm of compilers or domain-specific compilers, contributing to robust, high-performance systems.
  • Knowledge of Essential Tools: Demonstrate your command of vital development tools such as Linux, Git, GCC, and LLVM, ensuring smooth, efficient workflows and high-quality results.
  • Adaptability in a Startup Environment: Thrive in our fast-paced, dynamic startup culture where your ability to take initiative and achieve impactful results is highly valued and rewarded.
  • Familiarity with Machine Learning and Deep Learning: Leverage your expert knowledge of machine learning and deep learning, with hands-on experience in popular frameworks like TensorFlow, PyTorch, Llama.cpp, and vLLM, to drive groundbreaking advancements in AI technologies.
Responsibilities
  • Successfully develop and deliver a functional AI compiler (either ahead-of-time or just-in-time) along with an associated runtime environment, ensuring robust performance and reliability.
  • Lead the Ampere Computing AI compiler software efforts, driving innovation and excellence in Ampere’s compiler technology.
  • Work hand-in-hand with the silicon design team to enhance the Ampere Aurora AI accelerator, contributing to cutting-edge hardware-software integration.
  • Design and implement advanced solutions and enhancements for prominent machine learning libraries, including Llama.cpp, PyTorch, JAX, vLLM, and ONNX (C++), significantly boosting their performance and capabilities.
  • Conduct in-depth analyses of neural network models, identifying and proposing optimizations at multiple levels including model architecture, framework efficiency, compilation process, and execution speed, leading to substantial performance improvements.

Ampere Computing specializes in high-performance processors for cloud applications, offering products that deliver superior performance while consuming less power. Their processors utilize a unique single-threaded core architecture and are scalable, making them suitable for various environments, from large data centers to edge computing. The Ampere Altra Family and AmpereOne Family cater to different performance needs, with a focus on sustainability. Ampere also provides developers with tools and resources to create cloud-native solutions on their processors.

Company Stage

Private

Total Funding

$369.6M

Headquarters

Santa Clara, California

Founded

2017

Growth & Insights
Headcount

6 month growth

1%

1 year growth

6%

2 year growth

31%
Simplify Jobs

Simplify's Take

What believers are saying

  • Ampere's 256-core processors align with the trend towards high-core cloud instances.
  • Partnerships with Qualcomm and SynaXG enhance Ampere's AI and 5G market opportunities.
  • Focus on sustainable computing meets rising demand for eco-friendly data center solutions.

What critics are saying

  • AWS's 896-core instance launch challenges Ampere's high-core processor competitiveness.
  • Oracle's low CPU core pricing pressures Ampere to reconsider its pricing strategy.
  • Rising AI accelerator prices could impact Ampere's cost structure and pricing strategy.

What makes Ampere unique

  • Ampere's processors offer 50% higher performance than Genoa and 68% than Bergamo.
  • Their single-threaded core architecture ensures high performance with lower power consumption.
  • Ampere's scalable design supports diverse environments, from data centers to edge deployments.

Help us improve and share your feedback! Did you find this helpful?

INACTIVE