Willard Khoa - AI Engineer Team Lead

Willard Khoa

Select me!

Willard is available for hire

Hire Now

Willard Khoa

Gender

Male

Main expertise

AI Machine Learning Deep Learning NLP Data Scientists Python Java C++TensorFlow PyTorch Computer Vision Image Processing Software

Introduction

Willard Khoa is a highly motivated and experienced AI Engineer with seven years of expertise in artificial intelligence. He is deeply committed to advancing the field and is skilled in Generative AI, Machine Learning, Deep Learning, and Natural Language Processing. With a proven track record of delivering high-quality solutions that drive business value, Willard is also a Kaggle Master with an impressive achievement of one gold, eight silver, and three bronze solo medals. He has demonstrated outstanding expertise in tackling complex problems and developing cutting-edge AI solutions. Proficient in a range of programming languages, including Python, Java, and C++, Willard has expertise in popular AI frameworks such as TensorFlow and PyTorch. He possesses a strong understanding of data structures, algorithms, and software design patterns, with excellent problem-solving skills and attention to detail. A collaborative team player with excellent communication skills, Willard is adept at working effectively with cross-functional teams to ensure project success.

Project list

Automated Review and Scoring System for Stock Photos and Footage

In a stock platform, maintaining the quality control of each item is crucial. This project aims to reduce the manual review workload of the labeling team by 50% through the application of AI technology, which automatically reviews and scores 100 million items.

Develop a scoring algorithm that accurately assesses performance based on the defined criteria. Integrate the scoring system into the existing platform or application. Develop techniques to identify content generated by AI models, such as analyzing language patterns, coherence, and consistency. Auto Score: Responsible for planning, development, and review. Auto Review: Implement generative AI content detection, abnormal event detection, and optical flow analysis. Implement a system to flag potentially generated content for further review. Utilize machine learning algorithms to detect unusual or anomalous events within the data. Monitor for patterns or deviations that may indicate fraudulent activity or system failures. Employ optical flow techniques to analyze video data and identify changes or movements. Detect abnormal behavior or objects that may indicate security threats or safety hazards.

Technologies: Business, Stock, Automation

Image-Based Search Integration

In many instances, users prefer searching for images with similar content rather than entering keywords. This project addresses that need by utilizing various techniques from the Image Retrieval domain, creating a feature that allows users to search using images. This functionality is seamlessly integrated into a stock platform's extensive database of 80 million images.

Oversee project planning, execution, and progress tracking, ensuring adherence to deadlines and budget constraints. Lead technical design and solution architecture, proposing and implementing innovative technologies like CLIP and Faiss to deliver a high-performance and accurate image-based search system. Develop and implement a deep learning-based image feature extraction model using architectures. Integrate the image search system with the stock platform's database of 80 million images, using APIs or data pipelines. Train and evaluate machine learning models for image feature extraction, indexing, and similarity search.

Technologies: Image Search, Information Retrieval

Developing a Machine Learning Testing Platform

This project is a collaborative effort between the AI Department and the Test Department, aiming to develop a comprehensive solution for testing AI systems across various use cases.

Management: Control plan and progress, discuss with HR Team Adversarial attack and defense (working with VNU-UET) Test set generation: Augmentation, Diffusion, GAN Robot grasping testing MLOps testing system

Technologies: AI Testing

Automated Test Scoring for Entrance Exams

This project aims to reduce the manual effort required by the HR Team during the recruitment process. It provides an internal tool to automatically evaluate test scores for every candidate, enhancing efficiency and accuracy in scoring entrance exams.

Management: control plan and progress, discuss with HR Team Template detection Checkbox detection Digit handwriting recognition Review code quality

Technologies: Internal Tool, Automation

Automated Electrical Diagram Converter

This project involves creating a solution for a Japanese electrical corporation that can automatically convert raw electrical sketches into drawing files compatible with existing electrical software. This innovation aims to significantly reduce the workload for electrical engineers.

Manage control plans and progress discussions with customers. Apply pseudo-labeling methods to reduce labeling effort and generate synthetic datasets for training. Develop and implement symbol detection and recognition. Review and ensure code quality.

Technologies: Engineering

OCR Integration in POS Systems for Discount Labels

Developed a system for Japanese supermarkets to integrate Optical Character Recognition (OCR) technology into Point of Sale (POS) machines. This system automatically applies discount labels to customers' items at checkout, enhancing efficiency and accuracy.

Design algorithms to filter low-confidence predictions Design an augmentation scheme to improve text recognition Review code quality

Technologies: OCR, Retail

AI Product Recognition Project

This collaborative project between the AI Department and the Test Department aims to develop a comprehensive solution for testing AI systems across various use cases.

Management: Control plan and progress, discuss with customers. Data filtering and generation: Synthetic data generation, Image Quality Assessment. Product detection: YOLO v7, DINO. Product recognition: EfficientNet, Swin Transformer, arcface. Review code quality.

Technologies: IRetail

Lightweight Object Detection for Autonomous Driving

This project focuses on developing lightweight object detection and classification models to address challenges in autonomous driving. Trained models are subsequently quantized and deployed on the MediaTek i500 board.

Object Detection and Recognition with Lightweight Models: - EfficientDet Lite - MobileNet - SSD - YOLO Quantization techniques: - Convert float32 to int8

Technologies: Embedded, Automation

akaDoc OCR Platform

akaDoc is a robust document digitization solution used across various industries, such as banking and insurance, to automate the extraction and storage of vital information from multiple types of documents.

Document Detection Retrieval and Classification Script Detection: Latin, Japanese, Chinese, etc. Text Recognition: Latin (Lite version), Japanese Optical Mark Recognition (Checkbox Detection)

Technologies: Document OCR

Project list

Education

Expertise