About Me
DevOps and release engineer based in Vilnius, Lithuania. In software engineering since 2009, from QA and test automation to DevOps; at Rocket Software since 2014, on build, release, and delivery automation since 2018. My current focus is enterprise agentic applications: AI that takes on real engineering tasks under controlled workflows.
Outside work I maintain a number of open source organizations and projects. More than 1,000 repositories on GitHub depend on cpp-linter alone, including projects at Apache, Samsung, Qualcomm, and Bloomberg.
Highlights
- cpp-linter, my C/C++ quality tooling: more than 1,000 repositories on GitHub depend on it, including projects at Apache, Samsung, Qualcomm, Bloomberg, LLNL, and Nextcloud
- Author of the Jenkins plugin Explain Error; member of the Jenkins GitHub organization
- Selected for the Anthropic Open Source Developer Program (2026)
- EuroPython 2025 proposal reviewer
- 240+ original technical articles since 2017 (blog, also published on WeChat)
Work Experience
Senior DevOps & Release Engineer | Rocket Software, Lithuania | Jul 2024 – Present
- Build enterprise agentic applications: AI agents on the GitHub Copilot SDK that take on engineering tasks, connecting code understanding, task orchestration, tool invocation, and feedback.
- Bring AI into everyday engineering steps such as code review and build-failure analysis.
- Own the team’s CI/CD and delivery infrastructure.
DevOps & Release Engineer | Rocket Software, Dalian | Jun 2018 – Jun 2024
- Led CI/CD transformation across multiple product lines: from manual and Bamboo builds to Jenkins with shared libraries, all build logic as code.
- Built infrastructure as code with Ansible for provisioning Jenkins and development environments.
- Containerized enterprise products: multi-arch builds with buildx, health checks, Kubernetes deployments.
- Designed the DevOps maturity badge system and scaled it to department level; drove Conventional Commits adoption across teams.
- Automated VM management with Jira + Python, adopted company-wide.
- Introduced code coverage reporting for multiple product lines.
- Multiple Rocket Build Awards; several solutions adopted into product roadmaps.
Software Engineer in Test | Rocket Software, Dalian | Dec 2014 – May 2018 — test automation lead for Rocket Discover: built the automation framework from scratch and led the team that automated its regression testing.
Software Engineer in Test | JD.COM, Beijing | 2012 – 2014 — automated testing and CI pipelines.
QA Engineer | SIMCOM (Shanghai) & Neusoft (Beijing) | 2009 – 2011 — test design and execution; led small QA teams.
Open Source Projects
cpp-linter — C/C++ code quality automation: clang-format and clang-tidy delivered as a GitHub Action and pre-commit hook. More than 1,000 repositories on GitHub depend on it, including projects at Apache, Samsung, Qualcomm, Bloomberg, LLNL, and Nextcloud.
commit-check — a policy engine for Git commit metadata: one versioned TOML policy validating commit messages, branch names, author identity, signoff trailers, and AI attribution, shared across local hooks, CI, and GitHub Actions.
conventional-branch — a branch naming specification and toolset; since 1.1.0 it natively supports AI Coding Agent branch prefixes.
Explain Error — a Jenkins plugin, in the jenkinsci organization, that uses AI to explain build failures, with AI Auto-Fix (automatic repair PRs), usage statistics and quota controls, and providers including OpenAI, Gemini, Azure OpenAI, DeepSeek, Qwen, Ollama, and AWS Bedrock.
Open Delivery Spec — an open specification and toolset for detecting, analyzing, and governing AI-generated code: AI-code detection, quality scoring, and policy gates that run as a CI quality gate (every PR on this blog’s repository goes through it).
More: devops-maturity (DevOps maturity assessment and badges), MkDocs NG (community maintenance of the MkDocs ecosystem), gitstats (Git repository analytics), jenkinsfilelint (Jenkinsfile validation without a Jenkins server).
Skills
- Engineering practice: CI/CD, build & release, containers (Docker), infrastructure as code (Ansible), code quality & coverage, software supply chain security (SLSA / SBOM)
- AI engineering: agentic application development, AI code review, GitHub Copilot SDK, LLM integration (OpenAI / Gemini / Bedrock / Ollama and others)
- Languages: Python, Shell, Groovy, Go
- Platforms: Jenkins, GitHub Actions, Linux, Windows, and enterprise environments including AIX
Community
- EuroPython 2025 proposal reviewer; attended PyCon Lithuania 2025 and published three days of conference notes
- Maintainer in the Jenkins and MkDocs ecosystems
- Anthropic Open Source Developer Program
Languages
- Chinese — Native
- English — Professional Working Proficiency
- Lithuanian — Beginner (A1)
Education
- College Degree, Software Technology — Liaoning Provincial College of Communications (2006 – 2009)
Contact
- Email: xianpeng.shen@gmail.com
- GitHub: @shenxianpeng
