之隐 上.

Key Skills

Software

LabelImg

CVAT

Doccano

Top Subject Matter

自动驾驶/道路场景 Domain Expertise

语音转写与说话人分离 Domain Expertise

人脸识别关键点标注 Domain Expertise

Top Data Types

Image

Text

Top Task Types

Object Detection

Transcription

Point/Key Point

Segmentation

Entity (NER) Classification

Classification

Freelancer Overview

高级数据标注专员（自动驾驶道路场景图像标注与质控）. Brings 7+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include LabelImg, CVAT, and Doccano. Education includes N/A, Guizhou Industrial Vocational and Technical College (2021) and Intermediate Professional Skills Certificate, N/A (2021). AI-training focus includes data types such as Image and Text and labeling workflows including Object Detection, Transcription, and Point.

Labeling Experience

Senior Data Annotation Specialist - Suzhou Futech Technology Co., Ltd.

CVATImageSegmentation

Served as a Senior Data Annotation Specialist leading automotive image annotation deliverables for computer vision applications. Managed end-to-end quality control and ensured annotation consistency across large-scale road scene datasets. Applied expertise in multiple annotation formats and tools to support production-grade dataset generation for downstream model training and evaluation. • Led 500k+ road scene image annotation including 2D/3D boxes and semantic segmentation • Developed and maintained 15+ annotation guideline documents and improved onboarding efficiency • Built a three-tier QC system (self-check, peer-check, random inspection) to reduce labeling error rates • Coordinated team workflow optimization and improved per-capita throughput by standardizing tool operations

2022 - Present

高级数据标注专员（自动驾驶道路场景图像标注与质控）

LabelImgImageObject Detection

主导自动驾驶道路场景图像的数据标注工作，覆盖2D框、3D框与语义分割等标注类型。通过制定并完善标注规范与验收标准，建立三级质检体系以保证标注质量稳定达98.5%以上。项目交付准确率达到98.7%，并提前2周完成交付以支持后续模型训练迭代。• 完成道路场景2D框、3D框与语义分割标注共50万+张图像• 建立自检-互检-抽检三级质检体系并降低错误率• 编写并优化LabelImg与Doccano高效操作手册提升团队产能• 负责12人标注团队的质量管控与进度追踪

2022 - Present

高级数据标注专员（智能客服NLP文本标注）

DoccanoTextEntity (NER) Classification

参与智能客服领域的NLP文本标注工作，完成客服对话数据的意图识别、实体抽取与情感分类。覆盖50+个业务场景并将标注数据直接用于智能客服模型训练。模型意图识别准确率从72%提升至90%，同时人工客服接待量减少40%。• 标注30万+条对话数据的意图、实体与情感信息• 覆盖50+业务场景以支撑模型泛化• 支撑智能客服模型训练与效果评估• 推动业务场景落地带来效率提升

2022 - 2022

Data Annotation Specialist - Yunshang Kunpeng Technology Co., Ltd.

CVATImageSegmentationClassification

Worked as a Data Annotation Specialist handling face keypoint, product image, and speech transcription annotation tasks. Ensured labeling accuracy and collaborated with leads on quality checks and mentoring for new hires. Utilized annotation workflows and guideline alignment to produce datasets supporting computer vision and NLP-related model development. • Completed 200k+ face images keypoint labeling to support face recognition training and iteration • Annotated 150k+ e-commerce product images with classification, attributes, and subject segmentation • Produced 800+ hours of speech transcription and speaker separation annotations with high transcription quality • Assisted with quality sampling reviews and trained 8 new labelers while supporting team performance

2020 - 2022

数据标注专员（人脸关键点标注）

CVATImagePoint/Key PointSegmentation

独立完成大规模人脸关键点标注工作，覆盖68点与106点两种关键点定义以支持人脸识别模型训练。通过持续对齐标注规范并确保准确率稳定，支撑模型迭代与产品落地。累计标注20万+张人脸图像，关键点标注准确率达98.2%。• 完成人脸关键点标注共20万+张图像• 覆盖68点与106点两套关键点标注方案• 支撑人脸识别模型训练与迭代验证• 按规范完成高一致性关键点标注

2020 - 2022

Show more

Education

Work History