diff --git a/programs/lfx-mentorship/2026/02-Jun-Aug/README.md b/programs/lfx-mentorship/2026/02-Jun-Aug/README.md index fc63257f..8dd2485b 100644 --- a/programs/lfx-mentorship/2026/02-Jun-Aug/README.md +++ b/programs/lfx-mentorship/2026/02-Jun-Aug/README.md @@ -86,10 +86,6 @@ Mentee application instructions can be found on the [Program Guidelines](https:/ - [KubeSlice](#kubeslice) - [KubeSlice Controller HA (Active/Standby) Support](#kubeslice-controller-ha-activestandby-support) - [Partial Mesh Support (MVP: Hub-and-Spoke)](#partial-mesh-support-mvp-hub-and-spoke) -- [KubeStellar](#kubestellar) - - [AI-driven bug discovery and remediation architect for KubeStellar Console](#ai-driven-bug-discovery-and-remediation-architect-for-kubestellar-console) - - [AI-driven operational knowledge base and mission control testing for KubeStellar Console](#ai-driven-operational-knowledge-base-and-mission-control-testing-for-kubestellar-console) - - [AI-driven test coverage architect for KubeStellar Console](#ai-driven-test-coverage-architect-for-kubestellar-console) - [KubeVela](#kubevela) - [LRU Cache Eviction for the Native Helm Provider](#lru-cache-eviction-for-the-native-helm-provider) - [Native Secret-Sourced HTTP Headers and CRD-Based Config Management with Server-Side Validation](#native-secret-sourced-http-headers-and-crd-based-config-management-with-server-side-validation) @@ -1008,46 +1004,6 @@ CNCF - KubeSlice: Partial Mesh Support (MVP: Hub-and-Spoke) (2026 Term 2) - LFX URL: https://mentorship.lfx.linuxfoundation.org/project/0daa13c2-8a4b-4dd0-8520-20f20de2e580 -### KubeStellar - -#### AI-driven bug discovery and remediation architect for KubeStellar Console - -CNCF - KubeStellar: AI-driven bug discovery and remediation architect for Console (2026 Term 2) - -- Description: The KubeStellar Console (console.kubestellar.io) has GA4 error tracking that surfaces JavaScript errors, failed API calls, and rendering failures in production — but no one systematically triages or investigates them. Common issues include undefined data guards causing crashes, stale WebSocket state showing outdated cluster data, chunk load failures on deployments, and silent error swallowing that hides failures behind blank cards. This project takes a novel approach: the mentee acts as a bug discovery architect, designing investigation strategies, analyzing production error data, and directing an AI coding agent (an advanced AI coding agent) to audit code paths, identify defects, and propose validated fixes with regression tests. The mentee is evaluated on bugs found, bugs fixed, reduction in production error rates, and quality of the automated bug-hunting workflows they create — not on lines of code manually written. Critically, the mentee builds autonomous GitHub Actions workflows that continuously find and fix bugs after the mentorship ends. This includes a weekly automated bug sweep that uses an AI agent to scan the codebase for new error patterns, unsafe access, and missing guards — auto-filing issues for each finding. A GA4 error regression workflow compares error rates week-over-week and auto-files issues when a category spikes. An auto-triage workflow triggers an AI agent to investigate new GA4 errors, reproduce them, and propose fix PRs. The mentee also builds a public Console Quality Dashboard tracking error trends, bug fix rates, and open issues, and presents findings at KubeStellar community calls at midpoint and end of term. Note: this program is independent from the companion "test coverage architect" program — this role focuses exclusively on production error analysis, bug discovery, GA4 data, and quality dashboards. It does not involve test infrastructure, coverage metrics, or CI pipeline work. The two programs have separate mentees, separate deliverables, and no overlapping scope. Prerequisite: access to and license for an advanced AI coding agent capable of autonomous multi-file code generation. -- Expected Outcome: Triage and root-cause analysis of all existing GA4 errors, >=25 bugs discovered/filed/fixed with regression tests, >=50% reduction in production GA4 error rate, weekly automated bug sweep GitHub Actions workflow using AI agent to scan for error patterns and auto-file issues, GA4 error regression workflow that auto-detects week-over-week spikes, auto-triage workflow that investigates new errors and proposes fix PRs via AI agent, public Console Quality Dashboard tracking error trends and bug fix rates, improved error states across all card types (blank cards replaced with actionable messages), documented bug-hunting playbook for future contributors, 2 community call presentations (midpoint + final). -- Recommended Skills: Analytical mindset for error triage and root cause analysis, basic familiarity with React/TypeScript and web error handling, experience using advanced AI coding agents, familiarity with GitHub Actions and browser developer tools, understanding of Kubernetes concepts helpful but not required. -- Mentor(s): - - Andy Anderson (@clubanderson, andy@clubanderson.com) - - Arpit Srivastava (@arpit529srivastava, arpitsrivastava529@gmail.com) -- Upstream Issue: https://github.com/kubestellar/console/issues/4190 -- LFX URL: https://mentorship.lfx.linuxfoundation.org/project/fa8a2bf8-3dba-4ce5-a0b1-ba9ee116a919 - -#### AI-driven operational knowledge base and mission control testing for KubeStellar Console - -CNCF - KubeStellar: AI-driven operational KB and mission control testing for Console (2026 Term 2) - -- Description: KubeStellar Console's mission control feature uses an AI assistant backed by a knowledge base (console-kb) to help users perform operational tasks — installations, upgrades, troubleshooting, and multi-cluster sync fixes. The knowledge base needs comprehensive, validated operational content, and the full mission control pipeline (user question → KB lookup → command generation → cluster execution) needs end-to-end testing to ensure generated commands actually work. The mentee acts as an operational knowledge architect, using an AI coding agent (an advanced AI coding agent) to generate runbooks, installers, YAML templates, and troubleshooting fixes, then tests the complete mission control execution loop against real clusters. Critically, the mentee builds autonomous GitHub Actions workflows that nightly re-validate all KB operational content against live clusters, auto-filing issues when content goes stale, commands fail, or prerequisites change. The mentee also builds a feedback loop tracking which user queries return no results or bad results, auto-generating drafts for missing content. The mentee presents their findings at KubeStellar community calls at midpoint and end of term. Note: this program is independent from the companion "test coverage architect" and "bug discovery" programs — this role focuses exclusively on operational knowledge content, mission control pipeline validation, and KB coverage. It does not involve UI test infrastructure, coverage metrics, production error analysis, or GA4 data. The three programs have separate mentees, separate deliverables, and no overlapping scope. Prerequisite: access to and license for an advanced AI coding agent capable of autonomous multi-file code generation. -- Expected Outcome: Runbooks covering all common operations (install, upgrade, rollback, disaster recovery, multi-cluster sync failures), library of validated YAML templates and fixes for known issues, end-to-end mission control pipeline tests (user query → KB → command generation → successful cluster execution), nightly GitHub Actions workflow that re-validates all KB content against live clusters and auto-files issues on failures, KB gap analysis with auto-generated drafts for missing content, >=90% of KB operational content validated as working on current KubeStellar versions, documented KB contribution guide for future contributors, 2 community call presentations (midpoint + final). -- Recommended Skills: Familiarity with Kubernetes operations (kubectl, helm, YAML), technical writing ability for operational documentation, experience using advanced AI coding agents, familiarity with GitHub Actions and CI/CD, understanding of multi-cluster Kubernetes concepts helpful but not required. -- Mentor(s): - - Andy Anderson (@clubanderson, andy@clubanderson.com) -- Upstream Issue: https://github.com/kubestellar/console/issues/4196 -- LFX URL: https://mentorship.lfx.linuxfoundation.org/project/a0fbeeef-f240-4c9f-8941-24a9c861001a - -#### AI-driven test coverage architect for KubeStellar Console - -CNCF - KubeStellar: AI-driven test coverage architect for KubeStellar Console (2026 Term 2) - -- Description: The KubeStellar Console (console.kubestellar.io) is the web dashboard for the KubeStellar multi-cluster platform. It currently has minimal automated test coverage — Playwright E2E tests exist but are flaky and not CI-blocking, and component-level tests are sparse. UI regressions, broken cards, and auth flow failures are caught manually or not at all. This project takes a novel approach: the mentee acts as a test architect, defining coverage goals, test plans, and acceptance criteria, then directs an AI coding agent (an advanced AI coding agent) to generate, iterate on, and validate the test suite. The mentee is evaluated on test design quality, coverage achieved, and bugs discovered — not on lines of code manually written. Work includes designing test matrices (browsers, screen sizes, demo vs live mode, error states), building Playwright E2E and Vitest component tests, stabilizing flaky tests to become CI-blocking, and — critically — building autonomous GitHub Actions workflows that continuously improve test coverage after the mentorship ends. This includes a workflow that detects untested new components in PRs and auto-generates test PRs via AI agent, a nightly test health workflow that detects flaky tests and auto-files issues, and a coverage regression gate that blocks PRs dropping below threshold. The mentee presents their testing strategy and results at KubeStellar community calls at midpoint and end of term. Note: this program is independent from the companion "bug discovery and remediation" program — this role focuses exclusively on test infrastructure, coverage automation, and CI pipelines. It does not involve production error analysis, GA4 data, or bug triage. The two programs have separate mentees, separate deliverables, and no overlapping scope. Prerequisite: access to and license for an advanced AI coding agent capable of autonomous multi-file code generation. -- Expected Outcome: Playwright E2E tests covering 15-20 core user flows, Vitest component tests for all 30+ card types, test matrix covering 3 browsers x 2 screen sizes x 2 modes (demo/live), all E2E tests stable enough to be CI-blocking, coverage dashboard with >=70% target, >=15 bugs discovered and filed, GitHub Actions workflow that auto-generates tests for new untested components via AI agent, nightly test health workflow that detects flaky tests and auto-files issues, coverage regression gate blocking PRs that drop coverage, documented test authoring guide for future contributors, 2 community call presentations (midpoint + final). -- Recommended Skills: Test design and quality assurance principles, basic familiarity with React/TypeScript, experience using advanced AI coding agents, familiarity with GitHub Actions, understanding of Kubernetes concepts helpful but not required. -- Mentor(s): - - Andy Anderson (@clubanderson, andy@clubanderson.com) - - Arpit Srivastava (@Arpit529Srivastava, arpitsrivastava529@gmail.com) -- Upstream Issue: https://github.com/kubestellar/console/issues/4189 -- LFX URL: https://mentorship.lfx.linuxfoundation.org/project/390bbb3a-e49d-4c1d-8b45-331f34a26e51 - ### KubeVela #### LRU Cache Eviction for the Native Helm Provider