Deep Learning Compiler Engineer
![Quadric](https://semiwiki.com/wp-content/uploads/job-manager-uploads/company_logo/2022/11/Quadric-Logo-SemiWiki-150x74.jpg)
Website Quadric
Role:
As a senior member of our platform software engineering team, you will be tasked with lowering and optimizing neural networks on the quadric EPU. You will design and implement algorithmic optimizations to extract maximum performance out of the Quadric architecture.
Responsibilities:
- Drive the lowering and optimization of cutting edge deep neural networks using Quadric’s technology
- Apply your skills and expertise in mathematical & algorithmic optimization toward solving NP-hard problems
- Collaborate within the software team to develop algorithms that optimize graph-based execution on the Quadric architecture
Requirements
- MS or Ph.D. in Computer Science, or related field, with a minimum of eight years of experience in the industry
- Strong background in numerical and/or algorithmic optimization
- Understanding of building application-appropriate heuristics for NP-hard problems
- Knowledge of both classical as well as ML algorithms, e.g., Computer Vision, DSP, DNNs, etc.
- Strong background in graphs and related algorithms
Nice to haves:
- Proficiency in C++ >= 11
- Experience using / developing in TVM
- Knowledge of front-end and back-end compiler techniques
Expected Outcomes in 12 months:
- Develop a deep understanding of the hardware platform and low level software and leverage that for optimal performance of applications.
- Have a proven track record of implementing optimization passes for efficient lowering of deep learning and high performance computing algorithms on the Quadric EPU parallel processor.
Benefits
- Provide competitive salaries and meaningful equity
- Provide a politics free community for the brilliant minds who want to make an immediate impact
- Provide an opportunity for you to build long term career relationships
- Foster an environment that allows for lasting personal relationships alongside professional one
Intel’s Death Spiral Took Another Turn