Opens intel.wd1.myworkdayjobs.com in a new tab
About This Role
- The Intel Neural Compressor team develops state-of-the-art model compression technologies, including quantization, pruning and sparsity, knowledge distillation, and low-precision training and fine-tuning.
- Our work spans algorithm research, product feature development, performance optimization, and contributions to the open-source community.
- We are looking for a highly self-motivated engineer to join our team.
- Key Responsibilities: Develop Intel Neural Compressor and its core algorithm tools, including AutoRound, and optimize them for Intel AI platforms such as CPUs, GPUs, and AI accelerators.
- Research and implement quantization and compression techniques for large language model (LLM), vision language model (VLM), and generative models, including text-to-image, text-to-video and world models.
- Track and explore emerging directions in efficient model deployment, inference acceleration, and fine-tuning acceleration.
Qualifications
- Qualifications: Bachelor’s or master’s degree in Computer Science or a related field.
- Solid understanding of deep learning, deep learning frameworks, and large language model (LLM) fundamentals.
- Familiarity with model compression techniques such as quantization and pruning.
- Proficiency in Python, C++, or other programming languages commonly used in deep learning development.
- Strong teamwork mindset and collaboration skills.
- Good verbal and written English communication skills.
Nice to Have
- Strong self-motivation, ownership, and problem-solving skills.
- Passion for technological innovation and practical engineering, with a commitment to continuous exploration and improvement.
- Experience in model fine-tuning, inference optimization, or related tool development is preferred.
Sourced directly from Intel’s career page
Your application goes straight to Intel.
Opens intel.wd1.myworkdayjobs.com in a new tab
Specialisation
Open roles at Intel
627 positions
Job ID
/job/PRC-Shanghai/AI-Frameworks-Software-Engineer---Model-Compression_JR0286404
Get matched to roles like this
Upload your resume once. We’ll notify you when matching roles open up.
Join talent pool — free