For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

Invite to Job
P
Pan A.

Pan A.

密语IM智能通讯系统(AI增强)测试经理:LLM应用测试用例生成与Prompt回归评测

China flagN/A, China

Key Skills

Software

Don't disclose
Other

Top Subject Matter

Ai测试(智能客服/llm应用、prompt工程与安全评测) Domain Expertise
Ai辅助测试用例生成与风险覆盖度评估(web3钱包/交易风控) Domain Expertise

Top Data Types

TextText

Top Task Types

Prompt + Response Writing (SFT)Prompt + Response Writing (SFT)
Red TeamingRed Teaming

Freelancer Overview

密语IM智能通讯系统(AI增强)测试经理:LLM应用测试用例生成与Prompt回归评测. Brings 11+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Don't disclose and Other. Education includes Bachelor of Engineering, Hebei University of Science and Technology (2016). AI-training focus includes data types such as Computer Code, Programming, and Text and labeling workflows including Prompt + Response Writing (SFT), Evaluation, and Rating.

Labeling Experience

Test Manager - Sotong Softon

TextTextRed TeamingRed Teaming

Managed quality assurance for an AI-enhanced distributed instant messaging system serving a large-scale user base. Designed and improved an AI-assisted test workflow that accelerated test-case generation and improved defect localization using large models. Also led functional, automation, performance, and observability practices for traditional IM testing to maintain reliability at scale.• Built an AI-assisted test case generation and log-analysis workflow using an LLM-based platform• Created LLM application test strategies including evaluation metrics and regression pipelines• Implemented AI safety and adversarial testing such as prompt injection and information leakage checks• Delivered traditional IM testing covering automation regression, capacity testing with JMeter, and online monitoring.

2025 - Present

密语IM智能通讯系统(AI增强)测试经理:LLM应用测试用例生成与Prompt回归评测

Don't disclosePrompt + Response Writing (SFT)Prompt + Response Writing (SFT)

在AI增强的智能通讯系统中,搭建并运营以大模型为核心的测试辅助平台与评测体系。使用Claude基于需求文档自动生成正常/异常/边界测试用例,并结合日志分析加速Bug定位与复现。针对智能客服与摘要/多轮对话模块设计测试集与评估方案,形成可用于回归的Prompt迭代验证流程。 • Claude生成测试用例并提升编写效率 • LLM应用模块测试集构建与对话一致性评估 • BLEU/ROUGE与人工打分的双轨评估 • Prompt版本变更后的回归流水线与效果对比

2025 - Present

Web3去中心化钱包高级测试工程师:AI辅助风险用例生成与覆盖度分析

Other

在去中心化钱包项目中,作为测试负责人构建面向资金安全的测试体系,并引入AI辅助增强风险测试覆盖。利用AI进行风险用例生成与异常交易场景的覆盖度分析,以提升风控相关测试的完整性与有效性。与链上交互抓包及SQL账务一致性校验结合,支持交易准确性与高可用验证。 • AI辅助风险用例生成 • 异常交易场景覆盖度分析 • 与Charles抓包定位链上交互问题联动 • 与SQL多数据源一致性验证联动

2024 - 2025

Education

H

Hebei University of Science and Technology

Bachelor of Engineering, Software Engineering

Bachelor of Engineering
2012 - 2016

Work History

S

Sotong Softon

Test Manager

N/A
2025 - Present
S

Song Hong Technology

Senior Test Engineer to Test Manager

Singapore
2023 - Present