Research Scientist, Machine Learning Efficiency

About the Role

Google Research Singapore invites accomplished experts to join as Research Scientists focused on computational efficiency in large-scale generative AI models, including large language models (LLMs), diffusion models, and generative video models. This role involves pioneering foundational work on algorithmic efficiency, model compression techniques, and accelerating inference processes that directly shape how next-generation AI technologies serve billions of users globally.

Within this role, you'll design and execute large-scale practical experiments, prototype algorithms, and create innovative architectures while managing project timelines and deliverables. Contributions extend to academia through publications and collaborations with global technical institutes and universities.

Key Responsibilities

• Develop advanced algorithms, sampling methods, and optimization strategies to enhance efficiency and flexibility in serving AI generative models. This encompasses model compression, quantization, and knowledge distillation approaches.
• Invent cutting-edge algorithms and LLM architectures aiming to boost computational efficiency and enhance the generalization capability during deep learning training.
• Enhance the full model deployment pipeline including novel formulations for pretraining, instruction tuning, reinforcement learning, and reasoning methodologies.
• Work closely alongside hardware and software teams to optimize inference engines and kernels tailored for diverse hardware and model architectures.
• Tune latency, memory bandwidth, and workload management to maximize performance.

Required Qualifications

• PhD in Computer Science, related discipline, or equivalent hands-on experience.
• At least one scientific publication or submission in conferences or journals like CVPR, ICCV, NeurIPS, ICML, or ICLR.

Preferred Qualifications

• A profound understanding of transformer architectures.
• Demonstrated ability in generating new research concepts, from formulation to deployment in fast-evolving environments.
• Outstanding leadership skills and capability to communicate across multiple teams for collaborative work.
• Strong passion for deep learning, machine learning, computational statistics, and applied mathematics.

Additional Information

Applicants currently entitled to work in Singapore without need for visa sponsorship will be given hiring preference.

Google Research Singapore is a recent extension of Google’s global research network committed to cutting-edge innovation in intelligent systems and AI advancements aligned with Google's broad product ecosystem.

Google is an equal opportunity employer and provides accommodations for disabilities on request.