Jan 2022 – Jun 2024 · 2 yrs 5 mos · Remote

• Recruited as Java/Spring engineer, created AWS CDK custom resource to create PostgreSQL schema/role/user, recruited to SRE team to save Akamai to Cloudfront CDN migration project, after which unhappy about the fragmented ops and decided to start my own ondemandenv.dev
• Besides devops tasks like setting up Github workflow, Vault, Artifactory, LBs, SGs, kustomize and argo projects, different DBs, migration K8s jobs:
• Leading infra as code with AWS CDK, GitOps across multiple AWS accounts.
• Accomplished:
• Leading Azure Entra ID, SSO, IDP, OIDC, JWT, OAUTH, AWS IAC/Org integrations across multiple AWS accounts.
• Leading Share RDS, IPAM pool, NAT across VPCs with transit gateways across multiple AWS accounts.
• Leading Joining AWS events with application log/trace/metrics with Athena/OpenSearch across multiple AWS accounts.
• Leading CDN migration to Cloudfront with Athena partitioned query logs, Cf functions, Labmda@edge WAF ….
• Leading Database schema/role/user as infrastructure code.
• Leading Cdk8s generating K8s manifests as infrastructure code.
• Leading AWS Codepipeline and Github status sync.
• Leading legacy CFN/CDK projects migration.
• Resolving dependencies among hundreds of stacks across multiple AWS stacks.
• Resolving inconsistency in different test environments.
• Troubleshooting/resolving unknown/unexpected blackbox situations.
• Daily with AWS CDK/Cloudformation custom resources. Github App, Java/Kotlin Springboot. EKS, Athena, PostgreSQL, RDS cluster. Vault.
• Very familiar with ArgoCD, K8s CRDs. operators. Helm, Istio, Kong, Lighthouse CI, Opensearch, ... ...

Jan 2020 – Sep 2022 · 2 yrs 8 mos · New Jersey, United States

• Dev-Ops enterprise apps on AWS/ECS/EKS
• JAVA/Jbang,Kotlin/Kscript/Gradle/Selenium/Cucumber/JS for server automation/integration;
• Lead migrating legacy to productive AWS CDK/EKS/Serverless with across regions failover:
• Infra as code, GitOps/automation. CDK/Cloudformation+ArgoCD/CDK8s across regions deployment;
• Gradle/Typescript/CDK8s/Helm/Github API
• Environment on demand by Git branching
• logging/monitoring with CW/kube prometheus/Elasticsearch/timescaledb

Jun 2019 – Oct 2019 · 4 mos · New York City Metropolitan Area

• Infra as code with Bash/Python, Chef/Knife/Kitchen/Inspec/Ruby, Terraform/AWS, Jenkins and K8s. Managing Jenkins servers/agents, plugins, jobs and clusters, maintain configurations, dependencies and scalability, migrating legacy manual created Jenkins jobs to code, troubleshooting build failures.

Apr 2014 – May 2019 · 5 yrs 1 mo · Greater New York City Area · On-site

• Full cycle, from concept/requirements analysis, coding, testing to deployment/maintenance. Full stack, from Adobe AIR publishing to Android/iOS, NodeJS/Webpack/VueJS/Bootstrap to MySQL indexing/triggers, AWS ECS/Docker. Multi roles, Gradle script/plugins to automate most tasks, migrate/maintain infrastructures, merging code, setting alarm, mentoring others on JMX memory monitoring/IntelliJ tools integrations/debugging remote port 5005 ....
• Infrastructure As Code: lead migrating dev ENV to 1) IntelliJ/Teamcity/Youtrack/Upsource, Nexus3 for Maven/Docker repos for CI/CD; 2) Multi isolated ENVs for different purposes in different AWS VPCs, achieving server side high availability/scale-ability, helping team focus on debugging and testing with peace of mind.
• Automation: lead migrating fully automation with Gradle, 1) to generate client API code, test/deploy artifacts( docker image, even iOS IPA ), 2) create/verify resources( Lambda triggers, subnets, SGs, S3 bucket policies, even ElasticSearch mappings )etc. 3) providing unified/detailed phases/contexts/tools for team to debug, maintain code, achieving decreasing devops’ ambiguities, misunderstandings and mistakes, let team focus on business logic hence increase development efficiency dramatically.
• Build/deliver apps/features: Architecture/implement heavily with Java 8/Spring5/Boot2/Integration etc., implementing apps/services loosely coupled with REST/RPC/SQS , following Domain Driven Design, achieving easy auto tests, organized( even visualized with IntelliJ ) extendable apps running in AWS Lambda/ECS/EC2, with less code/risk, more productive development and confident deployments, delivered 1)SAML integration with Okta/OneLogin; 2) Reporting with Solr; 3) Resource processing, text search with ElasticSearch; 4) Resource uploading, migrating, video multibit encoding; 5) Content Sync with RDS Aurora triggering Lambda ...

Oct 2011 – Sep 2014 · 2 yrs 11 mos · Ann arbor, MI

• Major/principle developer of NIH project: http://projectreporter.nih.gov/project_info_details.cfm?aid=7888843&icde=0
• Abstract: https://www.frontiersin.org/10.3389/conf.fninf.2013.09.00082/event_abstract
• In Stockholm:
• Poster: https://drive.google.com/open?id=0ByZEgZN1-wa1RGRsalBBNGY2UWc
• Slides: https://drive.google.com/open?id=0ByZEgZN1-wa1c2JoN0hzZ25QdjQ
• http://www.neuroinformatics2013.org/documents/abstract_book_2013.pdf (find pubanatomy)
• In Apache showcase: http://flex.apache.org/community-showcase.html (find pubanatomy )
• COMPLEX BUT INTUITIVE UI, BLAZING FAST SEARCH WITH AUTO SUGGESTING AND SUMMARIZATION Indexing 23 Millions documents (100G index still auto updating every midnight )!
• See video tutorials: https://www.youtube.com/watch?v=FMJZOGCFRX8
• See training presentation session: https://youtu.be/XhpoCOPDSYM
• CI/CD: Subversion source change driving Maven build:
• Sonatype Nexus server as dependency server, Maven/FlexMojos build/deploy war into remote Tomcat server driven by only click or source change.
• Managed by Teamcity, building multi source branches corresponding to multi environments, merging source between different branches.
• Data processing Pipelines:
• MEDLINE citation text mining, indexing more than 23 million document citations referencing to different concepts with Spring Batch+PostgreSQL, analyzing/indexing into Solr server, for searching, term suggesting and aggregating, heavily use of analyzer/tokenizer/filter.
• Mouse brain 3D models generating, with volume data from Allen Brain Institute, I used Spring Batch+HSQL, VTK, to generate thousands of 3d models, I used marching cubes, then smoothing algorithm to make models look good, models are binary files stored into MongoDB for dynamical loading and query with Spring Data MongoDB
• Gene expression 3D model generating, use Spring Batch+HSQL download gene expression data from Allen Brain Institute, generate 3D models for each gene, store into MongoDB for dynamical loading and query.

Feb 2011 – Oct 2011 · 8 mos · Beijing

• Meetings with various clients and engineers, analyze business workflow, deliver multi-tier web applications designs/architectures, implementation plans, automation/manual test plans,
• Relational DB Schema implementation, application modularization, server API design, Client side architecture/modularization, Critical Integration tests( most typical test case ) designs;
• MySQL, sharding, RabbitMQ, messaging, Tomcat/Apache as web server.
• Adobe Flex UI application architecture design auto build with Flex-mojos.
• Training mentoring junior/mid-level developers on auto build, agile and unit testing.
• Report to department manager, allocating resource.

Jun 2008 – Nov 2010 · 2 yrs 5 mos

• Following SCRUM process, develop new features,fixing bugs;
• Maintain and improve development continuous integration system.
• Maintain and improve automated test.

Jul 2005 – May 2008 · 2 yrs 10 mos

• Independently and remotely accomplish complicated outsourcing projects for clients locating in North America, Europe and China. Projects are mainly for CMS, game, Video Player and so on with the base of Flash player, some of them are still making $: http://www.playforia.net/play/pixxes/

Jul 2002 – May 2005 · 2 yrs 10 mos · On-site

• Coding servlet/jsp for Japnese and HP in Beijing