● Design and build the “Post Purchase Upsells” product for upselling products on ecommerce website

post-transaction page (not launched yet)

● Design and build asynchronous reporting feature for 3rd party client

● Design and build email reminder feature for client 3rd party upsells items

● Design and build generic integration and management system for clients to enable upsells

Design graph schema and observability metrics for performance monitoring and analysis on graph algorithms, packages and products.

API load testing on cloud platform using jQuery, and e2e automation test using Selenium; troubleshoot, debug and fix build failures from daily regression build.

Develop and maintain the on-prem browser based products, with Python and docker.

Support the cloud Jupyter platform for graph algorithms application, with kubeflow on GCP.

● Develop and maintain the data pipeline for public records in real estate, including data syncing, ingestion, extraction, transformation and loading to database (ETL), with services deployed on K8S.

● Design and implement multiple new features to the listings, end to end from data providers to customers: transactions’ buyers, sellers, property tax information, and mortgages along with the transaction, with kafka service for data streaming, AWS RDS for data storage, and React for front-end visualization.

● Collaborate with other teams for their feature requests: with search team to design and implement owner names parsing, standardization, and permutation to enable searching properties by owners’ name; with AI team to calculate and build ownership information (eg. number of years), and owner occupancy status to contribute to the sales prediction model and find potential sellers for agents.

● Enrich property database by introducing property foreclosure data and NYC property tax assessor data from new data sources, including raw data quality and coverage analysis by postgres SQL language, schema design, pipeline design and implementation and output validation.

● Set up a beta environment and enable pre-merge testing on Jenkins, to reduce gamma and production breaks.

● Migrate legacy Python projects with EMR jobs on EC2 servers to Java microservices and kafka on K8S, with architecture redesign and enabling incremental loading, to reduce cost and improve the data freshness.

● Develop and maintain a website for ticket escalation system, including communications and management

● Maintain the backend data flow for the website, including automatically submitting Aether experiments, pushing Aether job results to Cosmos, pulling cosmos data periodically and sending aggregation emails

● Maintaining the User Interface for the website, and the database for all customer escalation records relevant data, including escalation basic properties, Aether job results information, etc

● Work as a vendor at client (Microsoft)

● Design and expand monitoring and alerting systems for real-time performance, accessibilities, availability and synchronization statuses from Outlook related databases.

● Develop and optimize 10+ monitors in the monitoring system, using C# and SQL query, to help find performance issues before customer impact.

● Redesign and re-organize existing monitors to decrease the service workload and develop unit tests to improve code coverage (92%).

● Bug fixes to reduce false alerts and improve operational health; and redistribute multi threads for monitors, reducing execution time by 80%.

● Develop Torus cmdlets, Powershell scripts for remote connections, querying data from databases and automate deployments.

● Resolve performance issues and customer escalations regarding replication latency and database availability.