Binance Accelerator Program - Data Scientist (AI Agent Quality)

Asia && Vietnam, Ho Chi Minh && Australia, Brisbane && Australia, Melbourne && Australia, Sydney && Taiwan, Taipei && Hong Kong

Full-Time

Mid (3-6 yrs)

Data & Analytics

Posted on January 6, 2025

This project is dedicated to developing and researching AI Agent decision-quality assurance and continuous improvement for compliance and KYC-related initiatives. Core work involves applying statistical and data-analytical methods to perform root-cause analysis of AI Agent decisions, as well as researching and developing model optimization algorithms to drive iterative project improvements.

Responsibilities

Participate in and conduct research on decision-quality analysis and continuous improvement of the project

Develop and research statistical algorithms and machine learning models; use SQL and Python to validate and build production applications of large language models (LLMs)

Extract and analyze data to validate hypotheses and deliver quantifiable, actionable conclusions with clear implementation directions

Assist in producing analytical reports and technical documentation; collaborate with the team to drive project progress

Requirements

Master's degree or above in a STEM field (CS, Statistics, Data Science, Applied Mathematics, EE, or related)

Python proficiency: able to independently write data processing and analysis scripts (pandas, numpy, etc.)

At least one end-to-end project where you drove the work from start to finish with a tangible deliverable

Statistics / ML fundamentals: solid understanding of statistical models, machine learning model training and classification evaluation metrics (precision, recall, confusion matrix, FP/FN, and their business implications)

English reading and writing fluency (all project documentation and code are in English)

Strong teamwork mindset and a sense of ownership/responsibility for the project

Experience using AI coding assistants (Copilot / Cursor / Claude / ChatGPT) on real projects with a verification mindset (not blindly trusting AI output)

LLM / Prompt Engineering / Agent project experience

SQL data extraction experience

Background in financial risk, KYC, compliance, or anti-fraud

If you are someone who can wrangle messy data into actionable conclusions, leverages AI for efficiency but never takes its output on faith, and operates with genuine ownership — a STEM Master's-level candidate - we want you!

Apply for this position