Shengjie L.

Key Skills

Software

Google Cloud Vertex AI

OpenCV AI Kit (OAK)

Data Annotation Tech

Dataloop

Roboflow

Other

Top Subject Matter

Manufacturing

Optical Engineering

Robotics

Top Data Types

Image

Text

Document

Top Task Types

Classification

Evaluation/Rating

Data Collection

Prompt + Response Writing (SFT)

Freelancer Overview

I have hands-on experience in data processing, image analysis, and AI-assisted system development, particularly in manufacturing and optical engineering scenarios. During my internship at Trina Solar, I participated in an automated photovoltaic module color-difference screening project, where I worked on image preprocessing, region segmentation, color value extraction, and classification logic using Python and OpenCV. I also collaborated on the development of a deep learning-based screening model to classify modules according to different color-difference levels. This experience gave me a strong understanding of the full data workflow, from raw data collection and preprocessing to feature extraction, quality evaluation, and model-oriented classification. Currently, I work as a Digitalization and Information Systems Engineer at Realman Robotics, where I use AI tools to develop production-related digital platforms, including an AI logistics platform, an AI after-sales maintenance platform, and an AI production data platform. My key strengths are my interdisciplinary background in optical engineering, manufacturing data, image processing, and AI application development. I am able to understand domain-specific technical data, structure complex information, and support the creation of high-quality datasets for AI training, evaluation, and business analysis.

Labeling Experience

Image Processing and Classification for PV Module Color-Difference Screening

ImageClassification

I implemented Python/OpenCV image-processing algorithms for module segmentation, color-value quantification, and color-difference analysis as part of an automated photovoltaic module screening project. I collaborated on deep-learning-based screening and classification models to assess color differences in PV modules. My work contributed to the development and training of classification models using labeled image data. • Developed segmentation, labeling, and analysis routines for photovoltaic module images. • Labeled and classified images based on color-difference thresholds. • Used Python and OpenCV for annotation and data preparation. • Coordinated with colleagues to optimize deep learning model accuracy.

2024 - 2024

Education

Work History