Mandarin-Speaking LLM Training & Evaluation Engineer

TRULYYY PTE. LTD.; Singapore; On-site; SGD 90,000 - 150,000; Full time

Job summary

TRULYYY PTE. LTD. is seeking a data/model engineer to develop, curate, and validate high-quality training and fine-tuning datasets for real-world ML applications.

You will translate complex reasoning and domain knowledge into clear, machine-trainable formats and run end-to-end training and optimization experiments. You will design evaluation criteria, benchmarks, and edge-case tests, analyze model outputs for reasoning quality and accuracy, and collaborate with Mandarin-speaking stakeholders and

Qualifications

• Degree in Computer Science, AI, Data Science, Mathematics, Statistics, Engineering, Physics, or a related quantitative field.
• Strong technical foundation in machine learning and modern LLMs, with practical exposure to training, fine-tuning, or evaluation workflows.
• Proficiency in Python and familiarity with standard AI/ML tools and libraries.
• High standard for experimental rigor, dataset cleanliness, documentation, and scientific reproducibility.
• Mandarin Chinese speaking ability is required to collaborate with Mandarin-speaking stakeholders and regional engineering teams in daily work.

Responsibilities

• Develop, curate, and validate high-quality training and fine-tuning datasets.
• Translate multi-step reasoning, abstract concepts, and expert knowledge into clear, machine-trainable formats.
• Execute training, fine-tuning, and iterative model optimization experiments.
• Design robust evaluation criteria, benchmarks, and edge-case test suites tailored to real-world deployment needs.
• Analyze model outputs for reasoning quality, accuracy, consistency, and edge-case behavior.

Skills

Python

ML fundamentals

Experimentation

Education

Bachelor's degree in Computer Science / AI / Data Science

Tools

ML libraries

Job description

TRULYYY PTE. LTD. is seeking a data/model engineer to develop, curate, and validate high-quality training and fine-tuning datasets for real-world ML applications.

You will translate complex reasoning and domain knowledge into clear, machine-trainable formats and run end-to-end training and optimization experiments. You will design evaluation criteria, benchmarks, and edge-case tests, analyze model outputs for reasoning quality and accuracy, and collaborate with Mandarin-speaking stakeholders and