We are seeking a Research Assistant to develop accelerated AI, machine learning, and robotics algorithms with a strong focus on computational efficiency, memory reduction, and energy-aware deployment. The role targets foundation models, including large language models (LLMs), vision-language models (VLMs), and vision-language-action models (VLAs).
Requirements
- Develop accelerated AI/ML and robotics algorithms that significantly reduce computation cost, memory footprint, and power consumption.
- Design and optimize efficient training and inference pipelines for foundation models (LLM, VLM, VLA) across different model sizes and deployment settings.
- Apply and advance model compression techniques, including quantization, pruning, knowledge distillation, low-rank adaptation, and related methods.
- Conduct algorithm-hardware co-design to enable efficient and accurate deployment of AI algorithms on robotic, edge, and heterogeneous computing platforms.
- Develop methods for efficient deployment of AI models on cloud and edge devices, considering latency, throughput, and energy constraints.
- Implement, evaluate, and benchmark accelerated models using rigorous experimental protocols.
- Contribute to research publications in top-tier AI, ML, and robotics conferences and journals.
- Collaborate with interdisciplinary teams spanning AI, systems, and robotics.
- Contribute to open-source codebases and reproducible research practices.
Benefits
- Access to research facilities and resources
- Opportunity to work with interdisciplinary teams
- Potential for publications in top-tier conferences and journals