Researcher in Model Architecture - Remote

Overview

The role involves conducting groundbreaking research in neural network architecture design to advance state-of-the-art models, with a focus on real-time multimodal intelligence.

In Short

• Conduct research in neural network architecture design.
• Design novel architectures for model quality and efficiency.
• Explore capabilities like statefulness and long-range memory.
• Investigate architectural decisions impacting model trade-offs.
• Develop frameworks for evaluating architectural innovations.
• Collaborate with teams to translate research into scalable systems.
• Deep expertise in architecture design required.
• Strong understanding of deployment in cloud and on-device.
• Proficiency in balancing quality, efficiency, and adaptability.
• Exceptional analytical and problem-solving skills needed.

Requirements

• Experience with advanced architectures like state space models.
• Knowledge of generative modeling paradigms.
• Proven research track record in top-tier ML/AI venues.
• Strong programming skills in PyTorch or TensorFlow.
• Experience with profiling tools for model performance.
• Prior research in state space models or efficient Transformers is a plus.
• Experience in designing architectures for multi-modal systems.
• Startup experience or rapid innovation track record preferred.

Benefits

• Lunch, dinner, and snacks provided at the office.
• Fully covered medical, dental, and vision insurance.
• Pension plan available.
• Relocation and immigration support offered.
• Inclusive and supportive company culture.