headerAvatar

TC. Lin

Ta-Chun Lin | TC.

ML Engi|

Intro Video
Profile

Know More

About Me

I love programming and technology! Aside from my industrial experience, I once ran my own software development business, JAStudio Technology (https://jastudio-tech.com/), for around four years. I have a strong background in full-stack software development and machine learning, with experience in training AI models for industrial defect detection and developing software systems. These projects required me to apply software architectural patterns like MVC and object-oriented design patterns such as the Factory Method. For me, a career is a life ambition, and together, we can achieve remarkable milestones.

Loading...
About

My Background

Education

Master of IT, Computer Science

Queensland University of Technology

Feb. 2023 - Aug. 2024

Bachelor of Science

The University of Queensland

Feb. 2015 - Dec. 2017

Awards

• 3x QUT Executive Dean's Commendation for Academic Excellence (GPA: 6.83/7.0, 98%)

• QUT Best Research Paper Awards 2023

• Selected Representative of Master of Information Technology.

Professional Work

Experience

Machine Learning Engineer

RayTracer

Nov. 2024 - Present

• Anomaly Detection & Forecast: Developed a scalable anomaly detection system for time-series and temporal sensor data (10,000+ features, TB-scale). Utilized predictive forecasting techniques to identify anomalies in real-time.

• MLOps Infrastructure: Implemented end-to-end ML pipelines using MLflow & Airflow, enabling both real-time and batch inference in production environments.

• ML Training Platform: Built full-stack solution (React / Flask) for real-time inference and automated hyperparameter tuning, with integrated CI/CD for deployment.

Software Engineer

Spingence Technology

Oct. 2020 - Jan. 2023

• Industrial AI Model Training: Reduced missed detection rate by 50% within accepted overkill rate, and finetuned models for optimal yield; tested SoTA models, including classification, detection, segmentation.

• AI Software Development: Created frontend and backend of an edge AI Trainer with PyTorch, Flask, React, and Azure, enabling clients to train, finetune, and deploy AI models; Incorporated deployment version control tags.

• Image Storage Software Development: Developed frontend and backend of an image storage platform with different user permissions to save, access, preprocess training images with Flask, MongoDB, and React.

Software Engineer

Sinon Corporation Business Groups, KSI Inc.

Jan. 2019 - Sep. 2020

• Government ERP System Development: Built with .NET MVC framework, MSSQL, and AngularJS, implementing hierarchical authority permissions and optimizing performance with SQL stored procedures; CRUD performed on over 100,000+ data entries per query; Deployed the system with MS IIS Web Services.

Software Engineer

JAStudio Technology (Freelance Software Business)

Jan. 2019 - Jan. 2023

• Freelance Software Business: Check out URL in 'Portfolio' section.

• Delivered Projects in AI and Software: Specified in Project Experience Page.

Check Professional Projects from CV

My works in AI, ML, Software

Portfolio

Industrial Projects

Listed on CV / Resume

Find out all professional industrial projects via my CV.

Medium Blogs

Persornal Information

As a passionate lifelong learner, I love sharing my expertise through articles regarding technology.

JAStudio Technology

Persornal Information

Aside from work, my team and I run a freelance business in software under JAStudio Technology.

QUT Best Paper Awards 2023

Research Paper Honours

Research Topic: PCA-Clustering: Relationship between Downtime Events & Attributes in Steel Manufacturing.

Fruit Vision

Artificial Intelligence

This project trains a computer vision model to identify 100 different fruit classes (transfer learning of EfficientNetB2) - 9.1M parameters.

Deep Learning Drowsiness Detections

Artificial Intelligence

Detects if a person is drowsy by Computer Vision. Implements YOLOv5 to custom-build a deep learning model. The demo allows the users to upload or use real-time webcam object detections.

Duck Segmentation

Artificial Intelligence

Deep learning model that implements YOLOv8 to segment ducks in images.

AI Image Centre Solution

Artificial Intelligence

This is an enterprise AI image storage and labelling platform solution, working similar to Google Drive. AI Image Labelling can be done on the same platform while seamlessly integrating to the AI training application.

AI Image Labelling Brush (Prototype)

Artificial Intelligence

This is a prototype AI image labelling 'brush' tool, allowing to upload, label, and download labelled labels of an images with coordinates.

AI Trainer - Industrial 4.0

Artificial Intelligence

Enterprise Computer Vision AI training, embeddable to industrial applications. Capable of managing large scale training data.

Airline Service Satisfaction Predictions

Artificial Intelligence

A comprehensive data science analysis of airline company satisfaction. A ML model was trained upon utilizing the analyzed insight, obtaining the score of >90%.

Robust & Reusuable Framework

Software: C# Framework

Designed an extensive, comprehensive, reusuable board game framework that can be widely reused. PDF user manual available.

Running Food Shopping Site

Software: E-Commerce

Fully functional food ecommerce site for a client in Taiwan.

Running Art Trading Site

Software: E-Commerce

Fully functional artworks shopping platform with auctions for a client in Taiwan.

AI-Powered ERP Geological System

Software: ERP

Enterprise ERP system with complex data flow and calculations.

Enterprise ERP Accounting System

Software: ERP

Enterprise ERP system with complex data flow and calculations.

Covid-19 Cases Tracking

Software: Web-App

This Covid-19 tracker was built with map and countries during the pandemic period, including number of confirmed/recovered/death cases.

BI - Bicycle Sales Analysis

Business Intelligence

Comprehensive Business Analysis for Bicycle Store.

BI - Retail Grocery Store Analysis

Business Intelligence

Comprehensive Business Analysis for Grocery Store.

More Projects...

AI & Cross-Platform Softwares

More undisclosed enterprise projects that need to stay confidential due to NDA contract in the field of AI Industrial Apps, ERP Systems, Cross-Platform Softwares.

My Abilities

Skills

Sharing Knowledges

Blogs

My Latest 10 Articles

thumbnail

Mastering LeetCode (From a Software Engine...

Mastering LeetCode (From a Software Engineer’s Perspective)

Thanks for clicking in! I must admit, the title...

thumbnail

NLP back to basics: Text Representation Mo...

Large Language Model is hitting the entire world, however, without understanding the basics, it’d be hard to step in...

thumbnail

ML Feature Engineering: Efficiently Findin...

This article continues from the previous topic:

thumbnail

ML Feature Engineering: Feature Crosses

When training an ML model, there are times when data features have complex relationships, making it...

thumbnail

Feature Normalization: The essential step ...

Imagine that you are training a Machine Learning model that analyzes how the height and weight of a person affect he...

thumbnail

ML Feature Engineering: Dealing with Categ...

Most of the algorithms in traditional ML, namely, algorithms based on statistical equations, work best with numerica...

thumbnail

Forward Propagation: The Neural Network Pr...

This article continues from Neural N...

thumbnail

Neural Network Architecture: Stepping into...

This article continues from:

thumbnail

Logistic Regression: The Gradient Descent

Logistic Regression: The Gradient Descent (Part 2)

This article continues from

thumbnail

Logistic Regression: The One-Layer Neural ...

Fun fact: Logistic regression can be considered a One-Layer Neural Network!

Read more on Medium