A Practitioner's Kit · v1.0

The Pipeline Is
The Product

Data pipelines for ML are not ETL pipelines. They are product systems with customers, a spec, a quality bar, and a lifecycle — and they are increasingly the part of the AI stack that determines whether a model works in production or only on benchmarks. This kit is a structured reference for product owners managing that system: a domain primer, thirteen capability cards drawn from SAFe 6 PO practice, a self-scored maturity rubric, and a downloadable agent pack you can deploy in Claude or ChatGPT to get hands-on help with your specific pipeline.

Browse The Capabilities → Download The Agent Pack
Built By
Dana Juncu · Senior PM, Data & AI
Last Updated
May 2026
Framework Basis
SAFe 6 PO · ML Systems Practice
01

A Data Pipeline For ML Is Not A Delivery Project. It Is A Product System.

The most consequential mistake a product owner can make in an ML data pipeline is treating it like a software delivery project — scope it, build it, ship it, close it. A labeling pipeline or a data curation system does not get to be done, because the model it feeds does not get to be done. The world keeps generating scenarios the training data has never seen.

The spec is the product. A labeling specification that defines what counts as a correctly annotated item is the closest equivalent to a software architecture document. Without a versioned, enforced spec, quality is unmeasurable and every review cycle becomes a renegotiation. The first question to ask of any pipeline is not "how much data do we have" but "how precisely have we defined what correct looks like."

The Mechanism, In One Line
Spec quality vs. model quality

If your quality measurement system is the model's downstream performance, you have measured the output and missed the input. By the time the model fails in production, the labeling decision that caused it was made months earlier.

The customers are ML engineers. Their failure modes — workarounds in training scripts, inconsistent model behaviour, evaluation gaps — are the most direct signal that the pipeline is not serving its purpose. Treating ML engineers as consumers of the pipeline rather than as co-designers is one of the most common structural errors in data platform product ownership.

13
Capability areas mapped to SAFe 6 PO responsibilities in this kit
4
Maturity stages from Ad hoc to Product-grade, scored across 8 dimensions
More data is not the same as better data. Volume without coverage just scales the problem.
A pipeline that treats itself as done is already behind the reality it is supposed to describe.

Why Data Pipelines Fail As Products

  • The spec is never versioned Guidelines exist but are not controlled documents. Annotators in different regions apply different interpretations. QA cannot catch inconsistency it has no reference to measure against.
  • Quality is measured retrospectively The first signal of a quality problem is a model failure in production. By then the labeling decisions that caused it are months old and the data is already in the training set.
  • ML engineers are not treated as customers The pipeline delivers volume. The training team works around the gaps. The gaps never become backlog items because the workaround makes the problem invisible.
  • Edge cases are perpetually deferred Coverage is measured in total labeled items, not in scenario diversity. The rare but safety-critical scenario is always one sprint away from being addressed, indefinitely.
02

Thirteen Capabilities For Running A Data Pipeline As A Product.

Each capability maps a PO responsibility area to its common failure mode and concrete outputs. Categories reflect SAFe 6 PO practice areas and ML data pipeline specifics. Click a card for the full failure mode description, what good looks like, and what to produce.

Filter
03

Where Does Your Pipeline Currently Sit?

Eight dimensions, each worth 0–2 points. 0 = not at all, 1 = partially or informally, 2 = systematically and reproducibly. Tap a cell to score yourself. Total out of 16. Honest scoring tends to matter more than a high number — the point is finding the structural gaps you did not know you had.

Dimension
What Good Looks Like
Score
1. Spec Quality
A versioned labeling spec exists, is actively maintained, and is the authoritative source for QA decisions.
0
2. Quality KPIs
Quality thresholds are defined upfront, measured consistently, and visible to all teams touching the pipeline.
0
3. Customer Feedback Loop
ML engineers provide structured feedback on data quality; their failure modes feed directly into pipeline priorities.
0
4. Supplier Capacity Model
Supplier capacity is modelled in sprint planning; onboarding for new scopes uses reusable templates.
0
5. Tooling as Product
Tooling decisions are owned by the PO, tracked in the backlog, and treated as compounding investments.
0
6. Edge Case Coverage
Rare and safety-critical scenarios are actively tracked; coverage gaps are backlog items, not known unknowns.
0
7. Data Provenance
Every labeled item can be traced to a spec version, annotator, QA pass, and delivery batch.
0
8. Stakeholder Alignment
Roadmap priorities are legible to non-specialist stakeholders; trade-offs are explicit and signed off.
0
Score Yourself Above
Tap each row's score cell to cycle through 0, 1, and 2. Your total updates as you score.
0 / 16
04

Drop This Into Claude Or ChatGPT To Get Help With Your Pipeline.

A Primed Brief, Not A Chatbot.

The agent pack is a single Markdown file containing the domain primer, all thirteen capability cards in structured form, the maturity rubric, and a set of guided prompts that walk a model through producing a prioritised pipeline health summary tailored to your specific context.

Paste it into a Claude project, a ChatGPT custom GPT, or any LLM that accepts long-context system prompts. The agent will ask about your domain, your internal customers, your pipeline's current state, and your most pressing problem — then produce a structured assessment grounded in this kit. A JSON version is also available for any integration that prefers machine-readable input.

  • Works as Claude Project knowledge or a system prompt
  • Domain-agnostic — automotive, banking, healthcare, logistics
  • SAFe 6 PO responsibilities mapped throughout
  • Outputs a prioritised 90-day path, not generic advice
  • Honest about what it can and cannot assess without context
Download .md Download .json
# Data Pipeline PO Agent Pack # v1.0 — drop into a Claude project or system prompt role: "pipeline-po-assistant" stance: "practitioner, SAFe 6, honest" context: - "domain-primer.md" - "capabilities.json" - "maturity-rubric.md" opening_question: "What is the pipeline for, who are the customers, and what is the most pressing problem right now?" do_not: - "produce a plan without knowing failure modes" - "recommend volume increases for quality problems" - "overload with all 13 capabilities at once" output_format: "pipeline-health-summary.md"