Research Data Scientist – APAC

This is a fully remote position, open to applicants in Hong Kong, +2 more countries.

📋 Description

• Perform original research in the field of generative AI and large language models as they relate to cryptocurrency and financial sectors, emphasizing reasoning under market uncertainty, the alignment of trading agents, multi-agent coordination, and scaling during test-time.

• Create and implement robust experiments, which involve hypothesis development, variations in models and training, RLVR-based reasoning strategies, systematic ablation studies, and drawing statistically valid conclusions.

• Innovate new post-training approaches and assessment frameworks for large language models used in cryptocurrency contexts.

• Assess the quality of chain-of-thought reasoning for market analysis, consistency in agent decision-making, and resilience against adversarial prompt injection.

• Investigate test-time scaling methods, including process reward models, self-consistency mechanisms, and planning based on Monte Carlo Tree Search.

• Utilize research techniques to enhance agent reasoning capabilities under ambiguous and rapidly changing market conditions.

• Monitor advancements in research from venues such as NeurIPS, ICML, ICLR, ACL, EMNLP, and other fields adjacent to cryptocurrency.

• Discover opportunities to implement cutting-edge techniques to address challenges faced by Binance.

⛳️ Requirements

• A Master's or PhD in Machine Learning, Computer Science, Mathematics, Statistics, or a related discipline; exceptional candidates with a Bachelor's degree and notable research achievements may also be considered.

• 0–5 years of research or industry experience in ML/AI; significant academic lab or research internship experience is highly valued.

• Profound knowledge of transformer architectures, dynamics of large language model pretraining, and post-training methodologies—including RLHF, RLVR-based reasoning model training, and alignment strategies.

• Expertise in Python and PyTorch.

• Strong mathematical background in linear algebra, probability theory, information theory, and stochastic optimization.

• Bilingual proficiency in English and Mandarin to facilitate communication with international partners and stakeholders.

• First-author publications at leading AI/ML conferences (NeurIPS, ICML, ICLR, ACL, EMNLP, AAAI) on core topics related to LLMs and agents—such as reasoning, multi-agent systems, RL applications for LLMs, post-training techniques, agent memory, or similar fields.

• Research or engineering experience within a prestigious AI lab or team.

• Experience in designing reward functions, developing training pipelines, or establishing evaluation frameworks for LLM-based agents, focusing on building core systems rather than merely utilizing existing APIs.

• An AI-native development workflow, employing tools like Claude Code, Cursor, or Copilot Workspace as the primary environment for development.

🏝️ Benefits

• Influence the future of finance.

• Collaborate with exceptional talent within a user-focused global organization that features a flat structure.

• Engage in challenging projects within a dynamic and innovative environment that offers autonomy.

• Succeed in a results-oriented workplace with opportunities for professional advancement and ongoing learning.

• Attractive salary and company benefits.

• Remote-first work-from-home policy (may vary based on role and team requirements).