Shape the next generation of intelligent agents by leading original research in core areas including planning, reinforcement learning, multimodal reasoning, and human-agent interaction. You will explore foundational challenges in agent grounding, reward modeling, and task evaluation, aiming to improve both the capabilities and reliability of AI systems.
What You’ll Do
- Lead independent research initiatives that span conceptual design, data collection, model development, and rigorous benchmarking.
- Develop novel methods in agent architectures, learning paradigms, and evaluation frameworks to push the boundaries of current AI systems.
- Work closely with engineering teams to transition experimental models into scalable, real-world applications.
- Validate advancements through systematic experimentation and comparison with state-of-the-art techniques.
- Share discoveries through publications at premier conferences such as NeurIPS, ICML, CVPR, and ACL, contributing to the global research dialogue.
What We’re Looking For
You have a PhD in Computer Science, Machine Learning, or a related discipline, or equivalent research experience. You’ve led research projects from idea to publication and have a proven ability to execute complex technical work.
- Deep knowledge in one or more areas: large language or vision models (pre-training, fine-tuning, post-training), reinforcement learning, LLM-based agents, computer vision, multimodal learning, or representation learning.
- First-author publications at top-tier venues such as NeurIPS, ICML, CVPR, or ACL.
- Strong implementation skills using PyTorch, JAX, or similar frameworks.
- Experience designing datasets, running large-scale experiments, and interpreting results to guide iterative improvement.
- A drive to explore high-impact research with practical implications and a vision for transforming how AI agents operate in real environments.
