Senior Staff AI Research Engineer – On-Device Language Intelligence

Senior Staff AI Research Engineer – On-Device Language Intelligence

Requirements

• PhD in C.S., EE or related fields or equivalent combination of education, training, and experience
• 10+ years of research experience in the fields of AI/NLP/ML
• Experience conducting research and shipping user facing products
• Experience in large language model (LLM), including Transformer model architecture, attention mechanisms, decoder only LLMs, SSM architecture
• Foundational LLM training experience is a plus, including data curation, distributed training, and hyperparameter tuning
• Experience in making LLM-based solution deployable on-device with small latency and memory (e.g., knowledge distillation) and on-device acceleration
• NPU optimization is a plus
• Experience in LLM alignment, instruction tuning, LoRA, Adapter, etc.
• Expertise in multi-step reasoning, planning, reinforcement learning (including RLHF), etc.
• Proficiency in deep learning frameworks such as TensorFlow, PyTorch, or similar
• Strong analytical and problem-solving skills, with a keen attention to detail
• Excellent written and verbal communication skills
• Demonstrated ability to work independently as well as collaboratively in a fast-paced research and development environment
• A strong product/commercialization deliverable experience is required

Responsibilities

• Conduct cutting-edge research and development of large foundation models (LLM, VLM, and Reasoning) for future, including model design, efficient model training, instruction tuning, prompt engineering, planning, action and related topics
• Collaborate with a multidisciplinary team of researchers, engineers, and domain experts to understand requirements, develop prototypes, and deliver robust solutions
• Conduct thorough evaluations and analysis of model performance, identify areas for improvement, and propose innovative solutions to enhance the overall quality and capabilities of large language models
• Generate creative solutions (patents), publish research results in top conferences (papers)

Preferred Qualifications

• Foundational LLM training experience is a plus, including data curation, distributed training, and hyperparameter tuning
• NPU optimization is a plus