[CNCF LFX Proposal] OptimizationJob CRD: Hyperparameter Tuning Engine for Kubeflow Trainer

[tariq-hasan]

CNCF Project

Kubeflow

Term

2026 Term 3 (Sep-Nov)

Program Name

OptimizationJob CRD: HPO Engine for Kubeflow Trainer

Program Description

Description

Hyperparameter optimization is critical for maximizing model performance, and Katib has long provided it through the Experiment CRD. But that CRD was built for broad use cases, including Neural Architecture Search and arbitrary workloads, which makes ordinary tuning verbose, hard to validate, and reliant on a stateful helper pod and database per experiment. The community has agreed to replace it with OptimizationJob, a resource focused solely on tuning TrainJobs. That work is half finished: the design is merged, the API is in review, and the push-based metrics foundation has shipped - but nothing yet watches these resources, launches trials, reads results, or picks a winner. This project builds that missing runtime.

Expected Outcomes

• Tuning controller: launches TrainJobs from the template, enforces trial budgets, injects hyperparameters, evaluates reported metrics, and records the best trial
• Stateless suggestion service: a short-lived, per-job companion that proposes values and is cleaned up automatically, with no database or persistent state
• Katib compatibility layer: translation letting the new typed API drive Katib's existing search algorithms unchanged
• Testing: unit, simulated-cluster, and end-to-end coverage, including failure paths
• Documentation: user guide, runnable examples, and migration notes for existing Katib users

Technologies

Go, Python, Kubernetes controllers, CRDs, HPO frameworks

Skills same as Technologies?

• Yes, the required skills are the same as the technologies listed above.

Required/Desirable Skills

No response

Mentors

Tariq Hasan | @tariq-hasan | mmtariquehsn@gmail.com | tariq-hasan
Andrey Velichkevich | @andreyvelich | andrey.velichkevich@gmail.com | andreyvelich

Upstream Issue URL

kubeflow/trainer#3562

Application Prerequisites

• Resume
• Cover Letter
• School Enrollment Verification
• Participation Permission from school or employer
• Coding Challenge
• Custom Prerequisite (fill in details below)

Coding Challenge URL

No response

Custom Prerequisite Name

No response

Custom Prerequisite Description

No response

Custom Prerequisite — File Upload

• Yes — completion of this task requires the mentee to submit a file.

---

LFX program: CNCF - Kubeflow: OptimizationJob CRD: HPO Engine for Kubeflow Trainer (2026 Term 3)

[github-actions]

🤖 LFX Proposal Validation

LFX program title (83/100 chars):

CNCF - Kubeflow: OptimizationJob CRD: HPO Engine for Kubeflow Trainer (2026 Term 3)

✅ All checks passed.

⚠️ 1 warning(s):

• LFX LFID/email verification is not yet automated. Please manually confirm mentor emails are registered on https://mentorship.lfx.linuxfoundation.org.

✅ Passed checks

• CNCF Project: Kubeflow
• Term: 2026 Term 3 (Sep-Nov)
• Program Name length: 52/69 chars available
• Program Description: 1405/3000 chars
• Technologies: Go, Python, Kubernetes controllers, CRDs, HPO frameworks
• Skills: same as Technologies
• Mentors: 2 listed (primary: Tariq Hasan)
• Upstream Issue URL: KEP-3562: OptimizationJob CRD for HPO: The Evolution of the Katib Project kubeflow/trainer#3562
• Quota: 3/4 proposals for Kubeflow in 2026 Term 3 (Sep-Nov)

What happens next

1. Maintainer approval — ⏳ a project maintainer (or authorized delegate) comments /approve.
2. Mentor confirmation — ⏳ 1 of 2 confirmed; @tariq-hasan auto-counted as proposer. Awaiting /confirm from @andreyvelich.
3. CNCF approval — ⏳ a CNCF admin comments /cncf-approve to finalize.

Proposal process guide

---

Re-run by editing the issue.