Chai C.

Key Skills

Software

Don't disclose

Other

Top Subject Matter

Computer vision (parking sign detection, object detection).

Computer vision QA and evaluation (food recognition).

AI workflow automation testing and evaluation.

Top Data Types

Image

Text

Document

Top Task Types

Object Detection

Fine-tuning

Freelancer Overview

Parking Sign Detection System (Computer Vision). Brings 13+ 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 Science, University Technology Sarawak (2014) and Bachelor of Applied Management, Federal University Australia (2011). AI-training focus includes data types such as Image, Computer Code, and Programming and labeling workflows including Object Detection, Evaluation, and Rating.

Labeling Experience

AI Engineer / n8n Automation Developer - CV Model Training

OtherImageFine-tuning

Trained and fine-tuned multiple computer vision models (YOLO, DenseNet, EfficientNet, ResNeXt) using PyTorch and TensorFlow. Applied transfer learning on a custom dataset containing 10K+ manually annotated images. Created data augmentation pipelines to increase robustness across varied input conditions. • Transfer learning and fine-tuning across YOLO and several CNN backbones.• Use of PyTorch and TensorFlow for model training pipelines.• Reliance on a custom labeled dataset (10K+ images).• Data augmentation to improve robustness and reduce overfitting.

2022 - 2026

AI Workflow Automation Test Framework

Other

Built an automated test framework to validate n8n workflow behavior and critical AI-driven automation paths. Implemented API endpoint validation, webhook trigger checks, data transformation accuracy checks, and error handling verification. Integrated data-driven testing via JSON fixtures into CI using GitHub Actions to ensure regressions are caught early.• Automated n8n workflow validation with Playwright + JavaScript.• API, webhook, data transformation, and error-handling test coverage.• Data-driven tests using JSON fixtures and automated reporting.• CI integration with GitHub Actions for repeatable validation.

2022 - 2026

Parking Sign Detection System (Computer Vision)

Don't discloseImageObject Detection

Built a custom image dataset for parking sign detection with manual annotation and labeling. Trained and fine-tuned object detection models using transfer learning on the annotated images. Developed augmentation pipelines to improve model robustness and generalization.• Dataset creation and manual labeling for 10K+ images.• Model benchmarking for accuracy and inference speed (YOLO, DenseNet, EfficientNet, ResNeXt).• Transfer learning fine-tuning using pretrained weights in PyTorch.• Data augmentation (rotation, scaling, flipping, brightness) for robustness.

2022 - 2026

Computer Vision QA System (SparkFood)

OtherImage

Validated and tested an image processing QA system for food recognition using vector database–backed semantic search. Ensured consistency of processed nutrition data by running test cases and checking expected versus actual outputs. Documented defects with reproducible steps and severity classification to support iterative improvements.• Test case execution for food recognition AI system.• Verification of image processing pipeline accuracy and semantic search results (Qdrant).• Consistency checks for nutrition data outputs.• Bug logging with reproduction steps and severity levels.

2019

Education

Work History