Binance is at the forefront of the global blockchain ecosystem, renowned for being the driving force behind the world's largest cryptocurrency exchange. Trusted by a vast global community of over 280 million users across 100+ countries, Binance is recognized for its unparalleled security measures, transparent user fund policies, swift trading engine, deep market liquidity, and diverse range of digital asset products. The company's offerings span across trading, finance, education, research, payments, institutional services, Web3 features, and more. Using the power of digital assets and blockchain technology, Binance aims to establish an inclusive financial ecosystem, promoting financial freedom and enhancing financial accessibility for individuals worldwide.

About the Role

We are looking for an accomplished Research Scientist/Engineer to drive advancements in the reasoning and planning capabilities of extensive foundation models. As part of this role, you will be responsible for improving model performance across various stages of development, such as data acquisition, supervised fine-tuning, reward modeling, and reinforcement learning. Your tasks will involve enhancing reasoning and decision-making, synthesizing high-quality datasets through rewriting and augmentation methods, solving intricate tasks with System 2 thinking, and utilizing advanced decoding strategies like MCTS and A*. You will devise and implement effective evaluation techniques, enable models to interact with external tools and APIs, and develop agents and multi-agent systems capable of tackling complex real-world challenges.

Responsibilities

Enhance reasoning and planning processes for foundation models throughout the development lifecycle to elevate overall performance.
Utilize rewriting, augmentation, and generation techniques to synthesize high-quality, large-scale datasets and boost the capabilities of foundation models.
Employ advanced reasoning strategies such as MCTS and A* to address complex tasks.
Establish and implement robust evaluation approaches to gauge model performance at different stages.
Enable foundation models to interact with tools, API interfaces, and code interpreters. Build efficient agents and multi-agent systems for solving intricate tasks.

Requirements

Proficiency in research background with a focus on RL, LLM, and familiarity with large-scale model training.
Strong grasp of data structures, fundamental algorithms, and fluency in Python or C++/Java programming languages.
Preferred experience in influential projects or publications related to RL, NLP, or Deep Learning.
Exceptional problem analysis and resolution skills, with the ability to address challenges in large-scale model training effectively.
Excellent communication and collaboration abilities, enabling exploration of new technologies within the team and fostering technological advancements.

Why Binance

Be a part of shaping the future within the top blockchain ecosystem globally.
Collaborate with exceptional talent in a globally-focused organization with a flat structure that prioritizes user needs.
Engage in unique, fast-paced projects with autonomy in an innovative work environment.
Thrive in a results-oriented workplace, offering opportunities for career progression and continuous learning.
Competitive salary and comprehensive company benefits.
Work-from-home flexibility (subject to the specifics of the business team's work nature).

Binance upholds its commitment to being an equal opportunity employer, recognizing the significance of a diverse workforce in achieving success. By submitting a job application, you acknowledge having read and agreed to the Candidate Privacy Notice provided.

Data Structures Jobs in Web3 in Thailand