Data Science Engineering Foundation

title: Data Science and Engineering Job Registry description: Authoritative job names, lifecycle classes, skill and specialist routes, and durable outputs for Data Science and Engineering Coach sessions

Data Science and Engineering Job Registry

Purpose

Use this registry before offering or selecting work. A job is a user-confirmed unit of focus inside one continuing coaching session. The registry routes work; it does not duplicate the methods owned by a skill.

Registry

Optional supporting routes in the registry are job-specific skill support and do not limit the separate segment-level Research eligibility defined by the RPI depth matrix.

RPI depth matrix

RPI is a user-directed depth route inside the confirmed job, not a job or lifecycle. Keep simple or adequately evidenced work on the primary route. No RPI segment auto-activates. Before each eligible segment, state the demonstrated need, purpose, expected artifact, expected interaction cost, limits, and direct path available if the user skips it.

Substantial delivery means multi-step code or artifact production whose dependencies, validation, or interruption risk make direct episodic execution unreliable. The owning skill first accepts the domain design. RPI then plans and executes delivery and reviews conformance without changing the job, class, owning skill, coach state authority, output root, or durable-write gate.

RPI Researcher may run only as an internal helper to an active rpi-research phase for one bounded source-gathering lane. Its return is not an independent job, invocation, state record, evidence authority, or recommendation. The active Research phase verifies selected sources and exclusively owns canonical evidence IDs, findings, recommendations, and the primary artifact.

The cross-cutting concerns below are evaluated in addition to the routes in this table. They are not listed per job, because they apply by trigger rather than by job name.

Cross-cutting concerns

Privacy, Responsible AI, and telemetry are evaluated across every job rather than being optional per-job routes. Evaluate each concern whenever its trigger is present, name the concern to the user, and route to its owning skill for the contribution that skill owns.

A cross-cutting concern always surfaces an observation and offers concrete choices. It never blocks a durable write, never edits an artifact silently, and never decides for the user. The durable-write scan gate remains the only condition that blocks a write.

When a concern triggers, state which concern applies and why, name its owning skill, offer the contribution that skill can make, and let the user choose whether to take it. When the owning skill is unavailable in the installed collection, say so and continue without improvising its authority.

Skill authority boundaries

The Data Science and Engineering Coach routes to the data science and engineering skills below, plus the cross-cutting skills above. Keep their authority separate even when one job loads more than one skill.

The two Responsible AI authorities are sequential, not alternatives. rai-standards provides risk framing and standards alignment and is evaluated first whenever an AI or ML system is in scope. rai-planner provides a scoped assessment of one surfaced risk and is reached only after that risk is identified.

problem-framing and experiment both route to experiment-design and stay distinct. problem-framing asks what kind of data-science problem a stated business outcome is, and completes without choosing one. experiment assumes a direction and selects what to test.

Problem-framing completion evidence

This job has two completion shapes and neither is an abandonment.

• Normal completion records the candidate problem classes, the grounded-versus- original reasoning behind each, unresolved discriminating questions marked unknown, and gaps carrying an answer holder and a coordinator. No candidate is selected, ranked, or scored.

• An unstated business outcome completes with a single outcome-definition gap, its answer holder, its coordinator, and an explicit statement that no candidate analysis was performed. Record it as completed work, not discarded-cleanly.

Completion returns control to the coach. A feasibility or experiment job starts only after a separate user-confirmed transition.

The two evaluation authorities are adjacent and must not be blended. evaluation-design covers systems whose output is a response, such as assistants and agents. ml-experimentation covers a trained model's predictive performance.

Selection rules

1. Present the relevant jobs with their class and expected output.

2. Ask the user to choose; do not infer or activate a job silently.

3. Record the confirmed job and its class using the session-state protocol.

4. Load the primary skill before job-specific guidance.

5. Load a supporting route only for the concern it owns, then reconcile the result into the session artifact list without transferring authority to it.

6. When a capability is unavailable in the installed collection, state the gap and offer an available route. Do not improvise a missing capability.

Routing boundary

A supporting route contributes within its own authority. The coach retains session identity, job selection, transition decisions, state mutation, pause and resume, durable-write gates, and completion choices. Loading a supporting skill never changes the active job without user confirmation.

The RPI depth matrix is the only job-specific Research and execution-routing authority. Every route remains subordinate to the active job's primary skill, lifecycle class, coach state, transition protocol, and durable-write gate.

Provenance

This registry is repository-original guidance licensed under CC BY 4.0. It does not reproduce or summarize an external standard.