Lorne Luo

Industry-leading provider of fuel analytics and field technologies, serving 20k+ fuel station worldwide, billions of IoT records ingestion per month. As a member of wetstock data engineering team, I architected and developed multiple internal data ETL processes and reporting portals critical to the scalability/efficiency/continuity of the broader SaaS platform. I introduced numerous best practices and standards to the team that increased the effectiveness and reliability of our delivered services.

• Modernised hundreds of outdated Python2 Cronjobs to Python3 Airflow cluster, unified 6 copies of untracked code base into one optimised repo
• Lead architect to convert wetstock analysis IP in Excel/VBS to a 10x better efficiency Java/Python-based AWS serverless service which includes Lambda, SQS, Kinesis, Glue, Redshift and Athena
• Restructured data lake on AWS S3, refined data mapping/validation logic and created data archiving plan
• Established and maintained a series of CICD pipelines leveraging from Jenkins, Github Action, Linux Bash, Powershell, AWS SSM and EventBridge to automate service’s deployment, data transfer and infrastructure provisioning
• Created a fully automated pipeline critical for duplicating the QA environment from production data fragments, significantly accelerating the dev and integration testing cycle
• Integrated Auth0 to reporting services in various languages including Java and C#, migrated user accounts and supported user management
• Maintained data ingestion Docker images across various operating systems and architectures and assisted clients in deploying them to fuel stations worldwide
• Managed and monitored 80+ Linux EC2 servers using Prometheus, Grafana, Icinga, Ansible, AWS CloudFormation, and Terraform
• Created Rest APIs and frontpage in FastAPI, Flask and Angular to support BI platform Yellowfin

A leading web development agency in Australia demonstrated by the 64 website design awards.

• Develop solution and implement high-value international web applications as prime back end developer
• Follow up end to end process of SDLC including BA, coding, testing, CICD and hotfix on cloud base environment
• Support the whole team as expert of Python and backend development
• Sustain and manage reusable libraries and maintain a standard initialization repository.
• Fire fighter for urgent deadline and thorny project
• Hosted an Opensource-Insight channel in team’s Slack

Provide large scale, high-availability complete IPTV package solution

• Dev/Ops for IPTV middleware which provide multimedia content management and delivery to multiple devices.
• Focused on seansirvice-side architecture design and programming.

EPS is a comprehensive petroleum industry service solution include real-time inventory statistic, alarm management and monitoring & reporting, backed with AWS based data pipline which process millions IoT data per day.

• Proposed and established a Airflow cluster with 7 nodes, replaced 600+ Python2 cronjobs to Python3 DAGs
• Integrated Auth0 authentication with 3 production services for 6 clients, migrated hundreds users
• Worked with data scientist to migrate Excel/VBS scripted analytical IP into modern Python project using Pandas/PySpark
• Maintain CI/CD pipelines which using Jenkins Github Action Shell Powershell Kubernetes Terraform Cloudformation
• Managed and monitored 80+ Linux EC2 servers, utilizing Ansible Puppet Icinga Grafana Prometheus Statd
• Designed and finished a centralized clients metadata management system on serverless architecture, provide both webui and API
• 10+ featured tech-oriented wiki pages and skill share presentations, below are some slides:
  • Advanced Git
  • TDD and How to

A Generative AI project to automatically generate videos, karaoke tracks, and subtitles for music, and publish them on YouTube.

• Fully automated process for audio downloading, subtitle generation, video rendering and Youtube publishing
• Used a series of Generative AI includes ChatGPT, Stable Diffusion, Real-ESRGAN, and Wav2lip
• Multi-media video processing in FFMpeg, UVR5, Whisper and Librosa

I leveraged Machine Learning's power to help pool shop's customer to read the value on pool test strips.

• Collect supervised data which include the accurate reading and photos of pool test stripe
• Annotate pool test stripes on images, ultilised YOLOv8 to traine a ML model to detect them
• Normalized the brighness and color balance and get the RGB value for each ingredients
• Trained 7 supervised Machine Learning models to predict the value of pool test strips

A weekly decision making system for ASX share stock investment which is using Monte Carlo stochastic approach to calculate return-risk balanced ranking.

• Tech stacks: Python, Kiness, Airflow, DynamoDB, Lambda, Redis, Django, TA-Lib, Pandas, Numpy, Matplotlib, Seaborn

Event-driven Forex trading platform which support tick-level price streaming, support pluggable strategy, instant notification, order history.

• Event-driven architecture with, which support tick-level price streaming and pluggable event handler
• Well defined interface which could replace Forex broker easily
• Leverage machine learning algorithms like a data scientist to develop trading strategy
• Tech Stacks: Python, EC2, Jupyter, TA-Lib, Pandas, Numpy, Matplotlib, Sklearn, optunity, keras

International official website of Monash University Low FODMAP Diet, including public site, blog, e-commerce, course enrollment system and recipe certification system.

• Start as contractor, finished 60% codes of this million dollar valued project in 6 months
• Implement a Two-way data sync between Django and Salesforce for eCommerce platform
• Coding for kernel logic of complicated recipe certification workflow
• Leverage finite state machine, auto programming to accelerate dev
• Concluded a life-cycle through Automated Model-Driven Development Methodology which reduced development resource 20%-40% for new projects

A reliable, functional and user-friendly Veterinary Telemedicine Platform for progressive veterinarians / specialists and customers.

• GraphQL(Graphene) + Celery + React architecture
• Applied Model-Driven Development Methodology which accelerated about 20% development time and significantly enhanced maintainability

Makes high-quality custom-made apparel from design to delivery.

• Fine-tune Django DRF API performance, reduced around 95% response time
• Finish complicated Custom Apparel Designer in React
• Implement solution to get real time callback from Salesforce sync middleware

An enterprise TV and digital signage application server which can be provided as an appliance or can be deployed in the Cloud as a SaaS offering..

• Finish distributed tv device management by Django + Celery
• Implement FFmpeg encoding for media files and EPG streaming process
• Create CI/CD pipeline with Jenkins, Docker and Salt-stack

Multi-tenants order and logistics tracing management system include customer, order, delivery, tracking and reporting.

• Fully featured E-commerce order management system by Django, Celery, Postgres and Vue.js
• Screenshots

Top 3 computer science university in China.

After 4 years study, I quited due to migrating to Australia.