# Ml Pipeline

> Ml Pipeline Copilot Skill

- Skill: `salihcantekin/ml-pipeline` (Agent Skill, multi-file: 4 files)
- Install (CLI): `npx skillmds@latest add salihcantekin/ml-pipeline`
- Raw SKILL.md: https://api.skillmd.com/api/skills/salihcantekin/ml-pipeline/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: DevOps & Infra
- Author: salihcantekin (https://skillmd.com/u/salihcantekin)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/salihcantekin/ml-pipeline

---

# Ml Pipeline Copilot Skill

## Purpose
On-demand Copilot skill for `data-and-ai/ml-pipeline`. Use this when the user asks for focused help in this domain.

## Invocation Triggers
- Requests to design, implement, refactor, review, optimize, or troubleshoot `Ml Pipeline` concerns.
- Tasks requiring detailed production-grade practices rather than generic snippets.

## Core Skill Focus
- robust architecture decisions
- production reliability
- security-first implementation
- test and observability completeness

## Execution Procedure
1. Clarify requirements and constraints.
2. Propose minimal-risk technical approach.
3. Implement with strong defaults for security, tests, and observability.
4. Verify with explicit checks and summarize trade-offs.

## Required Deliverables
- Concrete implementation guidance (not abstract recommendations only).
- Testing strategy (unit/integration/regression).
- Operational considerations (logging, metrics, rollout risk).
- Follow-up improvements for scalability and maintainability.

## Modern Engineering Defaults
- Use typed interfaces/contracts where possible.
- Prefer incremental migrations over big-bang rewrites.
- Require compatibility and rollback notes for impactful changes.
- Ensure security and observability are not optional add-ons.

## Acceptance Checklist
- [ ] Plan is technically sound and minimal-risk.
- [ ] Output is implementation-ready and not a placeholder.
- [ ] Validation steps are explicit and executable.
- [ ] Edge cases and failure paths are addressed.
- [ ] Trade-offs are clearly explained.

## Do Not
- Produce placeholder outputs.
- Skip edge-case handling.
- Omit validation strategy.