Data/AI engineer

• May 2023 - Present

Artificial Intelligence Engineer

• Nov 2019 - Jul 2021

[Recommendation Systems] • Developed and improved recommendation systems using iterative ML models (Logistic Regression / LightGBM / Deep learning). • Utilized Docker for end-to-end model delivery with AWS Sagemaker environment, resulting in a +5-10% increase in CTR with each iteration. [Computer Vision / Deep Learning] • Applied Deep learning models to hotel images to enhance user experience. • Employed ResNet for categorizing hotel images into 20 landscape/interior labels, achieving 99% accuracy in Top-5 predictions. • Utilized MobileNet to select thumbnails for hotels, resulting in a +3% increase in CTR. [A/B Testing] • Performed AB test analysis for car-hire and hotel search pages. • Achieved 10X growth in car-hire during the pandemic. • Built data dashboards in Tableau and Databricks for stakeholder reference. [ETL] • Designed and implemented data pipelines in Spark/Databricks. • Generated a dozen key metrics and features to train models and conduct AB testing analytics. Show less

AI Engineer

• Jul 2019 - Nov 2019

[Reinforcement Learning] • Applied Computer Vision and Reinforcement Learning techniques to develop an automated mobile app UI Quality Assurance system [Reinforcement Learning] • Applied Computer Vision and Reinforcement Learning techniques to develop an automated mobile app UI Quality Assurance system

AI Engineer

• Aug 2018 - Oct 2018

[Forecasting] • Developed a model for price estimation of logistics product lines • Utilized forecasting techniques to predict future prices based on historical data and market trends [Forecasting] • Developed a model for price estimation of logistics product lines • Utilized forecasting techniques to predict future prices based on historical data and market trends

Data Scientist

• Apr 2018 - Aug 2018

[Competitive Analysis] • Conducted a comprehensive campaign with competitive analysis • Utilized data scraping techniques to gather information on retail locations from competitors [Data Analysis and Presentation] • Analyzed the collected dataset to identify patterns and insights for the campaign team • Translated the findings into actionable recommendations to enable stakeholders to make data-driven decisions and optimize the product launch strategy [Competitive Analysis] • Conducted a comprehensive campaign with competitive analysis • Utilized data scraping techniques to gather information on retail locations from competitors [Data Analysis and Presentation] • Analyzed the collected dataset to identify patterns and insights for the campaign team • Translated the findings into actionable recommendations to enable stakeholders to make data-driven decisions and optimize the product launch strategy

Data Scientist

• Aug 2015 - Mar 2018

[Simulation] • Designed and implemented a Python-based simulation system for S.F. Express. • Leveraged SimPy to accurately model the conveyor system within the Distribution Centre. • Created an intuitive front-end interface, enabling business users to input site designs and parameters with ease. • Developed a robust back-end service using Flask, which stored simulation parameters and results for each iteration, utilizing SQLAlchemy for efficient model management. • The solution surpassed three competitors in terms of speed and adaptability, securing a 2-year contract with S.F. Express. [Credit Risk Modeling] • Designed credit scorecard workflows using Python to replace SAS. • Created feature selection, binning, and WOE calculation in Python. • Converted the logistics model into scorecards in Python. [Segment Analysis / NLP] • Delivered client segment reports for Hong Kong Jockey Club. • Created segment analysis using clustering and generated data diagrams and charts in reports. • Built sentiment analysis using Word2Vec on mixed Cantonese and English restaurant reviews. Show less