Large Language Model (LLM) Evaluation Engineer

Kuailu Software; Singapore; Presencial; SGD 90.000 - 150.000; Jornada completa

Descripción de la vacante

Kuailu Software is seeking an experienced evaluation engineer to build and maintain automated LLM evaluation pipelines. You will cover general, agent, and persona-based benchmarks, enabling one-click evaluation, historical comparisons, and regression testing.

You will deploy and run benchmarks like MMLU, C-Eval, HumanEval, GSM8K, MATH, and IFEval, while tracking metrics for BFCL, τ-bench, and GAIA. Strong Python and evaluation framework experience are essential.

Formación

• Bachelor's degree or above in Computer Science, Artificial Intelligence, or a related field.
• Familiarity with mainstream LLM evaluation benchmarks and frameworks, such as lm-eval-harness, OpenCompass, and EvalPlus.
• Strong proficiency in Python, with the ability to independently build evaluation pipelines covering model inference/deployment, batch evaluation, and results analysis.

Responsabilidades

• Build and maintain an automated LLM evaluation pipeline covering multiple dimensions, including general capabilities, Agent capabilities, and persona/role-playing. The pipeline should support one-click evaluation, historical result comparison, and regression testing.
• Conduct general capability evaluations using benchmarks such as MMLU, C-Eval, HumanEval, GSM8K, MATH, and IFEval, including benchmark deployment, execution, and results analysis.
• Conduct Agent capability evaluations, including setting up evaluation environments and tracking metrics for benchmarks such as BFCL, τ-bench, and GAIA.
• Design and execute persona/role-playing evaluation frameworks, covering metrics such as identity recognition, role compatibility, multi-turn stability, and style consistency.
• Record and analyze evaluation results from training runs, conduct comparative analysis and anomaly detection, and produce checkpoint evaluation reports.
• Conduct regular intermediate evaluations during the pre-training stage to track the evolution and improvement of model capabilities.

Conocimientos

Python

LLM evaluation

Benchmark frameworks

CI/CD integration

Data visualization

Educación

Bachelor's degree in Computer Science / AI

Herramientas

lm-eval-harness

OpenCompass

EvalPlus

vLLM

SGLang

Descripción del empleo

Job Responsibilities

1. Build and maintain an automated LLM evaluation pipeline covering multiple dimensions, including general capabilities, Agent capabilities, and persona/role-playing. The pipeline should support one-click evaluation, historical result comparison, and regression testing.
2. Conduct general capability evaluations using benchmarks such as MMLU, C-Eval, HumanEval, GSM8K, MATH, and IFEval, including benchmark deployment, execution, and results analysis.
3. Conduct Agent capability evaluations, including setting up evaluation environments and tracking metrics for benchmarks such as BFCL, τ-bench, and GAIA.
4. Design and execute persona/role-playing evaluation frameworks, covering metrics such as identity recognition, role compatibility, multi-turn stability, and style consistency.
5. Record and analyze evaluation results from training runs, conduct comparative analysis and anomaly detection, and produce checkpoint evaluation reports.
6. Conduct regular intermediate evaluations during the pre-training stage to track the evolution and improvement of model capabilities.

Job Requirements

1. Bachelor's degree or above in Computer Science, Artificial Intelligence, or a related field.
2. Familiarity with mainstream LLM evaluation benchmarks and frameworks, such as lm-eval-harness, OpenCompass, and EvalPlus.
3. Strong proficiency in Python, with the ability to independently build evaluation pipelines covering model inference/deployment, batch evaluation, and results analysis.
4. Familiarity with LLM inference frameworks such as vLLM and SGLang, with the ability to deploy models for batch inference and evaluation.
5. Experience in evaluation data analysis and visualization.
6. Detail-oriented and rigorous, with a strong focus on ensuring the reproducibility and reliability of evaluation results.

Preferred Qualifications

1. Experience with Agent evaluation, particularly BFCL, τ-bench, GAIA, or SWE-bench.
2. Experience with persona or role-playing evaluation, such as CharacterBench or RMTBench.
3. Experience with evaluation automation and CI/CD integration.
4. Understanding of model training workflows, with the ability to understand the relationship between training checkpoints and evaluation results.