Lian Wang Email and Phone Number

Independent consultancy in design and development of cloud based distributed apps and real-time data services, and in ML data assembly, model creation, training, and ML Ops pipelines engineering.

Lead engineer/manager of Viome Backbone Team developing critical backend services in AWS that deliver industry leading healthcare products based on gene sequencing and ML models to consumer and health provider markets.

Undertook a mega project to design and built a massively scalable, multi-tenant distributed data service cluster in AWS that provided generic data services for: 1) dynamic data schema registry, 2) real-time data pub/sub collection and exchange, 3) near real-time analytics and biz intelligence.

Led Internet Marketing Science Team in building machine learning models at massive scale for audience segmentation and bidding strategies that are applied to paid search and social media channels globally. Pioneered the team's distributed machine learning methodologies and infrastructure that have more than tripled the team's productivity in model training and deployment iterations. Specialized in realtime big data processing and business support system.

2015/6: The integrated realtime data services platform officially replaced Google Analytics completely. Added support for subscription channel to persist data in Google Cloud Storage. Added automated sharding to backend storage to optimize i/o efficiency and economy. 2015/4: Added support for TCP-based subscription to the realtime data service platform, allowing website and personalization teams to receive and use actionable intelligence to drive for customer experience in sub-second response time.2015/2: Organization wide adoption of the realtime data services platform goes viral. Peak volume surpassed one billion events per day. All clients are completely self served with zero downtime.2014/10: Designed and built a 360-degree realtime data services platform. Using Google's BigQuery columnar backend, this platform collects, validates, and publishes any enterprise data and analytics.

2014/02: Built a RESTful web service to ingest realtime data using a linearly scalable backend.2013/11: Built a RESTful web service engine to serve a virtual data mart that could host virtually unlimited number of datasets from any data source (including mysql, SQLServer, Vertica, BigQuery, HBase, and Hadoop/HDFS.) 2013/08: Developed and productized Z3, the auto-encrypting Hadoop compatible Distributed File System abstracted over AWS S3. Its built-in 256-bit client-side encryption makes the 3rd-party cloud storage as secure as any locally guarded secrets.

Led the data services team to build the core analytics platform for digital marketing using Hadoop, Hive, MongoDB, and MySQL. It ingested and processed daily billions of records from 800+ websites and social media forums. The whole production stack was deployed in AWS cloud.

Led the Data Warehouse team to build big data solutions as well as traditional RDBMS-centric data services. Designed and implemented a Hadoop MapReduce-based web analytics platform that processed multi-terabytes of daily incremental data from Disney's 500+ websites.

Led the warehouse team to implement and maintain reporting and analytics solutions for a leading online music download and subscription business that serviced and paid royalties for over 100 million plays a month.

Made major contribution to developing a business intelligence system that managed keyword bidding on prominent search engines such as Google and Overture (Yahoo and MSN.)

Designed Getty's 2nd generation pricing and licensing model and implemented its OLTP db system. A key achievement was to reduce a 3-dimension mathematical model to 2-dimension for simplicity and superior performance.

Developed web based project management and team collaboration server apps.

Developed SqlServer database as OLTP backend for soft-switch and other telecommunication services.