Wei Hou Email & Phone Number

Sunnyvale, Ca, Us

Sunnyvale, Ca, Us

Seoul, Seoul, Kr

Search ML infra

San Francisco, Ca, Us

HBase performance and stability tuning, led the efforts to migrate Airbnb's HBase clusters from HDFS to S3. Design and implemented the system to replay online database to offline data warehouse in realtime.Led the effort to migrate Airbnb's Flink platform from Yarn to Kubernetes.Led the effort to create the config-driven framework for Spark/Flink processing that's used by all realtime users at Airbnb

Bentonville, Arkansas, Us

Search Indexing: Develop, architect and maintain both batch and streaming index pipeline for Solr index generation of 'Walmart.com', use Spark and Kafka for streaming processing, Avro, HDFS and Hive to store and keep track of serialized data, Cassandra to maintain the status of streaming updates and ElasticSearch to log and keep track of Kafka event life cycle. Data Pipeline: Develop, architect and maintain ETL and ML pipeline, uses Hive/MapReduce/Python to extract, attribute and combine data from user click stream, catalog item infos , customer reviews and other data feeds to generate both item-based and query-based data which is used for Search ranking. Data quality: data migration,maintain and optimize the whole Big Data eco-system of Online Grocery. create critical metrics on top of click stream and item attributes for data validation. Big Data system health monitoring and Walmart production issue support. Business Solution: working closely with business to transfer business requirements into data-driven solutions, cooperate with different teams to provide optimized E2E solutions.Other streaming pipelines: Migrate and maintain Walmart sponsored products (WPA) pipeline/item online & local store availability (OOS) pipeline. Use Spark and Kafka for streaming processing, Couchbase to store output as key-value pairs to support real time query.

New York, Ny, Us

■Explain complex concepts in small study groups and hold office hour for individual student discussion■ Hold and organize online discussion programs

Use Python to develop a web crawler, then use the web crawler to collect movies’ information from millions of pages by usingsuitable regular expressions.■ Develop and maintain the database to store all the information■ Design and improve the database for fast researchACADEMIC PROJECT:Project: Massive Data Analysis of NYC Taxi Data; Collect accurate NYC taxi data of 2013, use Hadoop and Pyshp to divide complex data into various neighborhoods and time intervals■ Compare taxi data with possible correlated data such as weather and census and put forward hypothesis of causation■ Fit taxi data and correlated data sources into several classic machine learning models and give explanations of the result

Master Of Science (Ms), Computer Science

B.E, Computer Science