- Experience
- 3+ yrs
- Salary
- —
- Openings
- 1
- Posted
- 1 month ago
- Work mode
- In office
- Education
- Master's degree
- Resume
- Required to apply
Where you'll work
Sign in to tell us what does and doesn't work for you here — it sharpens every match we show you.
Job description
About The Team
Sea Group is launching a new strategic AI division dedicated to harnessing the power of generative AI for transforming human connections, self-expression, communication diversity, and social interactions. The team is building a comprehensive Model-as-a-Service (MaaS) platform and a multilingual AI ecosystem across multiple countries. This department aims to develop pioneering Southeast Asian multilingual AI models and innovate AI-based applications.
The AI application group specializes at the crossroads of social interaction and artificial intelligence. Their mission is to utilize large language models (LLMs) to create digital personas acting as personal assistants and social connectors. Operating with startup agility but backed by strong corporate resources, they aspire to reshape human-AI interaction.
Job Description
- Design and develop multi-faceted evaluation metrics tailored to varied business needs.
- Perform red-team testing and safety validation by simulating complex and extreme user scenarios to identify issues like bias, hallucinations, policy violations, and misaligned values and initiate corrective actions.
- Analyze root causes of flawed model outputs and collaborate with algorithm engineers on prompt improvements and fine-tuning.
- Build automated evaluation pipelines using LLM-as-a-Judge techniques to accelerate product iteration efficiency.
- Research user psychology in social and companion applications, converting subjective experiences into measurable metrics to drive continuous product experience improvements.
Requirements
- Demonstrated experience enhancing automated evaluation systems, including classic methods, LLM-as-a-Judge frameworks, and automated flaw detection.
- Practical experience in improving automated data generation workflows such as Self-Instruct, Evol-Instruct, and Synthetic Data Reinforcement Learning.
- Solid software engineering skills with proficiency in Python, Shell scripting, and Linux operating environments.
- Master's degree or higher in Computer Science or a related discipline, with a minimum of 3 years of professional experience.
Preferred Qualifications
- Previous experience in AI evaluation, quality assurance automation, or NLP algorithm development at a major internet company.
- Extensive use or development experience with popular AI social products.
Minimum education
Master's Degree