LFX Mentorship 2026 Term 3

The Kubeflow Community is participating in the LFX Mentorship Program 2026 Term 3 (September - November). This page aims to help you participate in LFX Mentorship with Kubeflow.

What is LFX Mentorship?

The LFX Mentorship Program is run by the Linux Foundation and CNCF to help developers begin their open source journey. Mentees work with experienced project maintainers on real-world projects and receive a stipend upon successful completion.

For more information, see the CNCF mentoring repository and the LFX Mentee Guide.

How can I participate?

Thank you for your interest in participating in LFX Mentorship with Kubeflow!

Key Dates

Here are the key dates for LFX Mentorship 2026 Term 3, the full timeline is available on the CNCF mentoring page:

Steps

1. Review the Term 3 timeline on the CNCF mentoring page.
2. Join the Kubeflow Slack:
   • NOTE: please do not reach out privately to mentors, instead, start a thread in the #kubeflow-contributors channel so others can see the response.
3. Learn about Kubeflow:
   • Read the Introduction to Kubeflow
   • Review the Architecture Overview
   • Consider trying out Kubeflow
4. Review the projects below to decide which ones you are interested in.
5. Apply through the LFX Mentorship portal with following the prerequisites listed on respective LFX project.

Projects

Project 1: Abstracting Pod Lifecycle Diagnostics for Kubeflow Pipelines

Components: kubeflow/pipelines

Mentors: Alyssa Goins, Matt Prahl

Details:

Improve the Kubeflow Pipelines (KFP) user experience by surfacing Kubernetes pod lifecycle failures directly in the UI. The project spans the frontend (TypeScript), backend (Go), Kubernetes APIs, and Argo Workflows to make debugging much easier for ML engineers.

Skills Required/Preferred:

• Go
• TypeScript
• Kubernetes and pod debugging experience
• Kubeflow (preferred)

Links:

• LFX Mentorship Project
• Proposal Issue: cncf/mentoring#1928

Project 2: Evolve SparkClient into Kubeflow’s Unified Data Processing Layer

Components: kubeflow/sdk (SparkClient), kubeflow/spark-operator

Mentors: Shekhar Rajak, Tariq Hasan, Rishabh Singh

Details:

Extend SparkClient with observability, scheduled and streaming execution, deeper Kubeflow integration, and end-to-end Spark → Trainer workflows, making it the unified data-processing layer within Kubeflow.

Skills Required/Preferred:

• Python, Java
• Apache Spark
• Kubernetes
• Distributed data processing and ML workflows
• DataFusion/Arrow and Prometheus/observability tools (optional)

Links:

• LFX Mentorship Project
• Proposal Issue: cncf/mentoring#1975

Project 3: OptimizationJob: HPO Engine for Kubeflow Trainer

Components: kubeflow/katib, kubeflow/trainer, kubeflow/sdk

Mentors: Tariq Hasan, Aniket Shaha, Akshay Chitneni, Andrey Velichkevich

Details:

Build the next-generation Hyperparameter Optimization runtime for Kubeflow Trainer by implementing the OptimizationJob controller, Katib compatibility, stateless suggestion service, metrics processing, and production-quality testing.

Skills Required/Preferred:

• Go
• Python
• Kubernetes controllers and CRDs
• HPO frameworks

Links:

• LFX Mentorship Project
• Proposal Issue: cncf/mentoring#2013