Yang (Andy) Yuan Email & Phone Number

JAD Data is a data engineering team in JR Academy to develop a product recommendation system based on real retail data from Instacart. The project constructs an automated data pipeline based on various AWS services and other tools and platforms such as DBT, airflow and snowflake enabling data storage, transformation, integration, analysis, and ML model deployment. Responsibilities:• Collaborated with other two Data Engineers and one DevOps Engineer, and organised daily stand-up and sprint meetings, and monitored progress using Scrum.• Planed, Configured, Deployed and Operated ETL/ELT pipelines for raw retail data.• Developed machine learning models for clean retail data.• Deployed machine learning models to a public web page which served as the recommendation system interface.Achievements:• Successfully processed the dataset with 3 million customer orders on Instacart by constructing ETL pipelines with various AWS services, such as S3, Redshift, Glue, Athena, Lambda, SNS and EventBridge. • Conducted Pyspark programming in AWS Glue to transform and integrate raw data.• Successfully implemented Apache Airflow to create, schedule and monitor the workflow and wrote Airflow DAGs in Python to deploy the whole ETL process from scratch instead of manual operation. • Developed XGBoost model in SageMaker to predict the next product a customer would buy and achieved 0.84 in AUC for the validation dataset. • Produced a SageMaker Model endpoint and connected it to AWS API Gateway which communicated with the recommendation system interface. • Infrastructure as code: applied CloudFormation to build S3 and Glue Crawler.• Version controlled code in Github, and deployed and built stacks with Github Action. • Worked together with DevOps Engineer to achieved CI/CD with Jenkins.

Sydney, New South Wales, Australia

Suncorp Group is an Australian financial, insurance, and banking conglomerate featuring AAMI, GIO, Suncorp Bank, and Vero among its brands. With more than 9.5 million clients and 13,000 workers, it is a top 50 ASX-listed firm. The Pricing Queries and Validation (PQV) Team is in charge of analysing claim data and detecting product dangers in order to lower insurance costs and boost profitability.Responsibilities:• Collaborated with pricing analysts to explore the influences of autonomous technologies in vehicles on claim experience parameters, such as Claim Frequency, Average Claim Size and Cost Per Policy. • Created data pipelines for processing raw data of claim experience. • Conducted data analysis and modelling to reveal the underlying drivers of claim differences.• Proposed recommendations of modifications for the current pricing model of automobile insurance. • Reported to the senior stakeholders of the results of analysis. Achievements:• Collaborated with pricing analysts to extract data with 3 million claim records from Suncorp’s database. • Successfully built data processing pipelines in RStudio to conduct data quality check, data cleaning and transformation. • Successfully applied GBM model and Lasso model to conduct feature engineering. • Developed Zero censored Tobit Model to predict the claim experience parameters and achieved 0.92 in AUC. • Obtained Top 10 autonomous technologies that most influenced the claim experience by using the Tobit Model.• Produced a summary report of the main findings from the Tobit Model and presented to the head of PQV Team.

Foshan, Guangdong, China

Country Garden is a property development company based in Guangdong, China. In 2020, it is ranked 147th in Fortune Global 500 List. Country Garden features a market capitalization of over US$29.84 billion as of 2018, with more than 200 high-end township developments throughout China, Malaysia and Australia among its vast international project portfolio.Responsibilities:• Developed strategies of land investment in Guangxi Province to support the growth of Country Garden. • Collected and analysed the data for the investment evaluation reports, economic feasibility studies and risk assessment reports. • Reported periodically to the senior stakeholders to illustrate recommendations of site or project investments and expected return based on financial analysis. Achievements:• Successfully purchased 5 projects worth RMB$300 million in Guangxi Province on behalf of the company. • Successfully created dashboards of land and project information for stakeholders using Excel and Tableau. • Generated 36 investment reports.

Mülheim An Der Ruhr, North Rhine-Westphalia, Germany

Siemens AG is a German multinational conglomerate with headquarters in Munich and branch offices all over the world. It is Europe's largest industrial manufacturing firm. The Power Diagnostics Center (PDC) is in charge of analysing data from their power plants around the world in order to respond quickly to real-time anomaly detection warnings that could result in an unplanned shutdown or downtime incident.Responsibilities:• Collected and analysed daily operating data from Siemens gas turbines to track the devices' performance.• Analysed the alarms generated by GT-AID, a Siemens-developed AI software that monitors and diagnoses faults in gas turbines on a daily basis. • Investigated and modified those built-in models in GT-AID which always triggered false alarms.• Cooperated with Software Engineers to update the code in GT-AID, with the goal of reducing false alarms and enhancing forecast accuracy.Achievements:• Statistically analysed a dataset with 30 thousand records of false alarms using Minitab. • Successfully proposed modifications of three built-in models in GT-AID and worked with Software Engineers to update the code in GT-AID, decreasing the number of false alarms by 70%. • Generated two summary reports about the mechanisms of malfunctions in gas turbines and presented the strategy of improving modelling accuracy in GT-AID to stakeholders.

Melbourne, Victoria, Australia

The Royal Melbourne Institute of Technology or RMIT University, is a public research university in Melbourne, Australia. It focuses on applied research as well as outcome-related research and consultancy services, and has extensive partnerships with government and industry.Responsibilities:• Participated in Australian Research Council (ARC) supported project Fundamental Investigation of Heat and Mass Transfer in Nanofluids and conducted theoretical and numerical research. • Built and programmed numerical models, and conducted numerical simulation of flowing nanofluids.• Scientifically graphed and analysed simulation data using OriginLab. Achievements:• Published 6 peer-reviewed journal papers, 5 conference papers. • Successfully proposed a numerical model and improved the prediction accuracy from 73% to 89% for the simulation of flowing nanofluids. • Successfully developed new programming codes of simulating flowing nanofluids for the computational fluid dynamics software ANSYS CFX.

Master Of Business Analytics, H2A (Distinction)

Key skills: •Programming: Python, R, SQL •Tools: Tableau, R Shiny •Data Warehousing & Big Data: ER Diagram, Star Schema, Dimensional.

Doctor Of Philosophy - Phd, Mechanical Engineering, Computational Fluid Dynamics

Technical skills: Ansys CFX, Fortran Thesis: Numerical modelling of bubbly flows in nanofluids with and without heat transfer.

Bachelor Of Engineering - Be, Energy And Power Engineering, 87/100 (High Distinction)