AI Software Solutions Engineer (AI Frameworks, Workloads)
Intel Corporation
Bengaluru, India
Job posting number: #7293153 (Ref:JR0267975)
Posted: November 5, 2024
Job Description
Job Description
Job Description
We are looking for a senior contributor to design, develop and optimize AI frameworks for Inference. In this role, you will work with a cross-geo teams to enhance the inference stack to ensure competitive performance on deep learning inference models with a specific focus on the PyTorch framework.
The roles and responsibilities that you would need to performance may include the following:
- Design and develop SW techniques for AI frameworks - both HW-agnostic and HW-aware
- Contribute to enhancing and extending the Inference and Training capabilities in our Software stack
- Profile deep learning inference workloads as needed and identify optimization opportunities
Qualifications
- BTech, MS or PhD in CS or related fields with an overall experience of 10 to 15 years
- Atleast 2 or 3 years of experience working on Inference frameworks/tools for inference for deep learning models and that have been deployed/used by customers
- Architecture/Design contributions to Inference systems
- Detailed understanding of machine learning systems optimization and deployment techniques such as quantization
- Experience with optimization techniques for deployment of Large Language Models (LLMs)
- Deep implementation knowledge of transformers and inference specific optimizations
- Programming skills in Advanced C++, Python and parallel programming skills
- Ability to debug complex issues in multi-layered SW systems
- Understanding of SW integration across open source frameworks and internal framework layers
- Strong understanding of computer architecture
- Effective communication skills and experience with working in a cross-geo setup
Preferred
- Experience working on and contributing to Inference serving solutions
- Knowledge of compiler algorithms for heterogeneous systems
- Knowledge of open source compiler infrastructure like LLVM or gcc
- Understanding of low-level kernels