Role Summary:
The Data Quality Analyst ensures high precision and quality for training and evaluation data essential for cutting-edge Multimodal Large Language Models (MLLMs). You will perform critical assessment and validation across diverse data types, directly contributing to the advancement of our AI's performance and intelligence.
Key Responsibilities:
- Multimodal Data Processing: Execute high-quality annotation, classification, and structuring of complex text, image, video, and audio data according to project specifications.
- LLM Output Evaluation: Strictly rating, ranking, and providing detailed feedback on LLM/MLLM responses (e.g., assessing accuracy, coherence, and safety).
- Quality Control & Guideline Management: Enforce strict quality assurance protocols, audit annotation results, and actively contribute to the refinement and iteration of data annotation guidelines.
- Cross-functional Support: Collaborate efficiently with engineering and data science teams, providing timely data insights and feedback for continuous model improvement.
Qualifications:
- Education: Bachelor's degree in Information Technology, Software Engineering or higher qualifications.
- Essential Skills: Excellent communication and collaboration skills; strong attention to detail and commitment to high-quality output; ability to quickly adapt to new rules and tools.
- Languages: Fluency in both Chinese and English is required as this role work in a globally team and collaborate with China team.
- Preferred Experience (Plus but not required):
- Prior work experience in data annotation or AI data preprocessing.
- Foundational knowledge of Deep Learning or LLMs; basic data development experience (e.g., SQL, Python scripting).
- Passionate about gaming, especially FPS (First-Person Shooter) or SRPG (Strategy Role-Playing Games), as the critical thinking and meticulousness from these genres are highly valued.
