Scientific Research Pipeline Migration Brief

Role: Principal Research Software Engineer specializing in reproducible scientific computing workflows.

Context

• Scientific research domain: {{research_domain}}
• Existing script ecosystem: {{legacy_script_environment}}
• Target orchestration framework: {{target_pipeline_framework}}
• Dataset scale and parameters: {{dataset_characteristics}}
• Reproducibility compliance standard: {{reproducibility_standards}}
• Migration target schedule: {{delivery_timeline}}

Task

Produce an actionable pipeline migration brief that translates legacy {{legacy_script_environment}} scientific workflows into a deterministic, containerized data pipeline on {{target_pipeline_framework}} satisfying {{reproducibility_standards}}.

Method

1. Inventory existing computational scripts across {{legacy_script_environment}} to identify state mutations and hidden dependencies.
2. Model the input/output artifact contracts given {{dataset_characteristics}} to define immutable data boundaries.
3. Map core analysis steps to discrete, idempotent tasks supported natively by {{target_pipeline_framework}}.
4. Design containerization and dependency isolation strategies to guarantee execution parity across environments.
5. Integrate verification checkpoints that validate results against {{reproducibility_standards}} after each processing stage.
6. Formulate error-handling and partial-retry protocols suited for long-running scientific simulations.
7. Construct a sequential transition roadmap aligned with {{delivery_timeline}}.

Constraints

• MUST eliminate implicit local filesystem state and hardcoded absolute paths.
• MUST NOT recommend proprietary runtime extensions that break open-science verification.
• Architectural designs must support compute throttling to avoid budget overrun.
• Technical decisions must stay within the scope of {{research_domain}} requirements.
• Maximum total length must remain under 600 words.

Output format

1. Architectural Blueprint

• Core migration flow and artifact boundaries (max 150 words)

2. Pipeline Execution Matrix

• Markdown table: Stage Name | Legacy Script | Target Task | State Management

3. Reproducibility & Validation Controls

• Checkpoint specifications matching {{reproducibility_standards}} (bulleted, max 100 words)

4. Phased Implementation Roadmap

• Milestones organized by {{delivery_timeline}}

Self-review

• Ensure every legacy script pattern in {{legacy_script_environment}} has an explicit containerized equivalent.
• Check that artifact contracts prevent silent data corruption on {{dataset_characteristics}}.
• Confirm no step requires manual runtime intervention.