Principal Engineer – GPU AI Kernel Library

Job Description:

AI software framework group is responsible for development of all the AI Frameworks to enable Intel’s AI HW roadmap. This group focus on developing AI software from kernel (such as matrix multiplication) to higher level frameworks (such as Inference engine). This group works closely with other Intel groups (such as lower-level stacks close to operating systems and higher-level stacks for cluster level scheduling) so that End to End software stack is developed for Intel hardware roadmap. This position, you will be responsible for prototyping, developing of GPU kernel, integrating to AI frameworks, and optimizing them for Intel hardware platforms.

Designs, develops, integrates, tests, validates, and/or debugs software to enable Intel product features to enable or utilize Artificial Intelligence, including machine learning and deep learning. Understands internal and external partner software and develops software across the stack (spanning firmware, drivers, OS, middleware, frameworks, algorithms, and applications) as required to enable and optimize specific AI features, capabilities, solutions, reference platforms, or Intel products. May include the development of reference AI software and improving or enabling customer designs to obtain the greatest value of Intel AI products, the development and/or optimization of workloads for AI benchmarks, and workloads for simulation to support preSi power and performance analysis and architecture recommendations. As a principal engineer, recognized as a domain expert who influences and drives technical direction across Intel and industry. Develops and mentors other technical leaders, grows the community, acts as a change agent, and role models Intel values. Aligns organizational goals with technical vision, formulates technical strategy to deliver leadership solutions, and demonstrates a track record of relentless execution in bringing products and technologies to market.

Qualifications:

Minimum qualifications are required to be initially considered for this position. Preferred qualifications are in addition to the minimum requirements and are considered a plus factor in identifying top candidates.

Minimum Qualifications:

• 3 Years experience -GPU software architecture
• 1 years experience – GPU Kernel Developments
• 5 Yrs of experience in complex C++ based development and meta level programming

Preferred Qualifications:

• AI kernel developments (such as flash attention, MoE)
• Understanding of overall AI SW stack
• Understanding of PyTorch architecture
• Ability to understand the complex requirements and lead large team of engineers with clear direction.
• Demonstrated to adhere to specific timelines and deliver quality SW product
• Requirements listed would be obtained through a combination of industry relevant job experience, internship experiences and or schoolwork/classes/research.