Join Meta as a Software Engineer focused on developing compiler systems for cutting-edge machine learning hardware designed for augmented and virtual reality applications. This is a unique opportunity to architect and implement a new compiler tailored for advanced deep learning technologies.
As part of our innovative team, you will collaborate with experts in compilers, machine learning algorithms, software, firmware, and ASIC design. Your contributions will directly impact the full-stack development process, transforming PyTorch models into executable binaries for specialized hardware accelerators.
Your Responsibilities:
- Analyze and design effective compiler passes and optimizations.
- Implement and enhance code generation aiming at machine learning accelerators.
- Collaborate with research teams to map ML graphs to hardware, analyze cost-benefits, and estimate power and performance.
- Contribute to the development of machine-learning libraries, intermediate representations, and analysis tools.
- Conduct design and code reviews, ensuring high-quality output.
- Evaluate code performance, troubleshoot issues, and lead resolution processes for compiler and system challenges.
- Improve the efficiency, scalability, and reliability of our toolchains.
- Work with other compiler teams to share and adopt innovative solutions.
Minimum Qualifications:
- Bachelor's degree in Computer Science, Computer Engineering, or a related field.
- A minimum of 2 years' experience developing compilers or code optimization tools.
- Proficiency in Python and/or C/C++ with a focus on software design and performance analysis.
- Knowledge of AI frameworks and their hardware implementations.
Preferred Qualifications:
- Familiarity with deep learning algorithms, especially convolutional and recurrent networks.
- Experience with active compiler toolchains, such as LLVM, MLIR, GCC, or Glow.
- Background in machine-code generation or compiler back-end development.
- Strong communication skills for effective cross-functional teamwork.
- Experience with mainstream machine-learning frameworks, such as PyTorch or TensorFlow.
Meta is dedicated to building technologies that foster connections and empower communities. By joining us, you will play a role in shaping the next evolution of social technology as we transition from traditional screens to immersive experiences.
Location: New York, NY