Hi Jenkins Community,
I’ve been following the GSoC 2026 project idea regarding an AI Chatbot to guide user workflows. After auditing the current AI in DevOps landscape, I believe the biggest hurdle for Jenkins users isn’t just writing code, but contextual troubleshooting.
Most generic AI models fail because they don’t know the user’s specific plugin versions or agent configurations. I am proposing a plugin designed to act as a Resident Expert.
Key Differentiators:
-
Local Context Awareness: The plugin will securely read build logs and controller metadata to provide specific fixes, not generic advice.
-
Sovereign AI Support: Recognizing the security needs of Jenkins environments, the architecture will support Local LLMs (via Ollama) so no data leaves the controller.
-
Pipeline Refactoring: Leveraging my experience in Zombie Code refactoring (from my Rosetta Zero project), I want to help users modernize old Scripted Pipelines into Cloud-Native Declarative ones.
As a contributor to IntelOwl, I’ve engineered automated threat intelligence orchestrators and implemented circuit-breaker architectures, proving my ability to handle complex plugin ecosystems. My work on Ayush Sentinel Guard involved building sovereign, privacy-first AI frameworks, which directly informs my strategy for bringing secure, local LLM capabilities to Jenkins. Additionally, through Rosetta Zero, I developed logic to refactor legacy Zombie Code into cloud-native structures, a skill I will apply to help Jenkins users modernize scripted pipelines into declarative ones. My proficiency in Rust, Python, and Java ensures I can navigate the Jenkins core and agent-side execution environments with ease
I’ve already looked into using the jenkins-core extension points for UI integration. I’d love to hear from potential mentors if the focus on Privacy + Local Context aligns with the community’s vision for this feature!
@krisstern @shivaylamba I’ve been analyzing the previous work done by Chirag Gupta on failure diagnosis and would like to build upon that by focusing on proactive workflow guidance. , I specifically have a question about whether a Local-First approach using Ollama fits the community’s resource constraints for the Jenkins controller?
Best regards,
Niyati Jain