Remote Foundation Models Scientist - Reasoning

Binance; Singapore; Remote; SGD 150,000 - 210,000; Full time

Benefits offered by this job

Remote-first

Work from home

Job summary

Binance is seeking a highly skilled Research Scientist/Engineer to advance the reasoning and planning capabilities of large foundation models. You will enhance model performance across data acquisition, SFT, reward modeling, and RL, while driving innovations in reasoning and decision-making.

The role involves solving complex tasks using advanced decoding strategies and teaching models to interact with tools and external APIs.

Qualifications

• Proven experience in RL, NLP, or Deep Learning.
• Fluency in Python or C++/Java.
• Strong problem analysis and collaborative skills.

Responsibilities

• Reasoning and planning for foundation models across data acquisition, evaluation, SFT, reward modeling, and RL.
• Synthesize high-quality data via rewriting, augmentation, and generation for pretraining, SFT, and RL.
• Utilize system 2 thinking and advanced decoding strategies (e.g., MCTS, A*).
• Develop robust evaluation methodologies to assess model performance.
• Teach models to use tools, APIs, and code interpreters.
• Build agents and multi-agent systems to solve complex tasks.

Skills

RL expertise

LLM knowledge

Python/C++/Java

Large-scale model training

Job description

Binance is seeking a highly skilled Research Scientist/Engineer to advance the reasoning and planning capabilities of large foundation models. You will enhance model performance across data acquisition, SFT, reward modeling, and RL, while driving innovations in reasoning and decision-making.

The role involves solving complex tasks using advanced decoding strategies and teaching models to interact with tools and external APIs.