Data architects and enterprise data engineers operate at the intersection of business strategy and technical execution — a demanding space where decisions made today can shape an organization's data infrastructure for years. ChatGPT can accelerate your workflow by helping you draft schemas, evaluate architectural trade-offs, communicate complex designs to non-technical stakeholders, and stay sharp in a rapidly evolving landscape. Whether you're designing a lakehouse from scratch or governing petabytes of production data, these prompts will help you think faster and build better.
Data Modeling & Schema Design
Prompt 1: Dimensional Model Review
You are a senior data architect. Review the following relational schema and suggest how to transform it into an optimized star schema for analytical workloads. Identify fact tables, dimension tables, slowly changing dimensions (SCD types), and any denormalization opportunities. Schema: [paste your schema here]
This prompt accelerates schema transformation work by giving you a structured, expert-level critique that surfaces SCD handling and denormalization trade-offs you might overlook under deadline pressure.
Prompt 2: Entity-Relationship Diagram Narration
Given the following list of business entities and their relationships, write a clear narrative description of an entity-relationship diagram (ERD) suitable for a technical design document: [list entities and relationships]
Converting stakeholder requirements into a communicable ERD narrative saves time in design reviews and ensures alignment before any DDL is written.
Prompt 3: Third Normal Form Audit
Analyze the following table definitions for violations of first, second, and third normal form. For each violation found, explain the issue and provide a refactored version of the schema: [paste table definitions]
Automating normalization audits catches design debt early and produces documentation-ready explanations that can be shared with junior engineers.
Prompt 4: Data Vault 2.0 Conversion
Convert the following source system entities into a Data Vault 2.0 model. Identify hubs, links, and satellites, define appropriate hash keys, and explain the load sequence. Source entities: [list entities and business keys]
Data Vault modeling requires strict discipline around naming conventions and load order; this prompt enforces that discipline and generates a reproducible starting point.
Prompt 5: Schema Evolution Strategy
I have a production schema that needs to evolve to support new business requirements without breaking existing consumers. Here is the current schema and the new requirements: [details]. Propose a backward-compatible migration strategy including versioning approach, deprecation timeline, and DDL steps.
Schema evolution is one of the highest-risk activities in enterprise data engineering, and a structured migration plan reduces the chance of costly downstream breakage.
Data Pipeline & ETL Architecture
Prompt 6: ELT vs ETL Trade-off Analysis
Compare ELT and ETL architectures for the following use case: [describe source systems, data volumes, transformation complexity, and target platform]. Provide a recommendation with justification, including cost, latency, and maintainability considerations.
This prompt produces a decision-ready comparison document that can be attached directly to an architecture decision record (ADR).
Prompt 7: Pipeline Failure Mode Analysis
Act as a data reliability engineer. For the following pipeline architecture, identify the top 10 failure modes, their likelihood, potential impact, and a mitigation strategy for each: [describe pipeline stages, technologies, and data volumes]
Proactive failure mode analysis reduces incident frequency and gives your team a runbook foundation before problems occur in production.
Prompt 8: Idempotent Pipeline Design
I need to design an idempotent data pipeline for the following process: [describe the pipeline]. Explain the patterns and mechanisms I should implement to ensure the pipeline produces the same result regardless of how many times it is re-run, including how to handle late-arriving data.
Idempotency is a non-negotiable property for enterprise pipelines; having a structured approach prevents data duplication bugs that are notoriously difficult to diagnose after the fact.
Prompt 9: Streaming Architecture Design
Design a real-time streaming data architecture for the following requirements: [describe source event streams, processing logic, latency SLA, and downstream consumers]. Include technology choices, partitioning strategy, state management approach, and fault tolerance mechanisms.
Streaming architecture decisions have long-lived operational consequences, and this prompt forces comprehensive coverage of concerns that are easy to defer until they become production emergencies.
Prompt 10: Data Lineage Documentation
Generate a data lineage document for the following pipeline. For each stage, describe the inputs, transformations applied, business rules enforced, and outputs. Flag any points where data quality could degrade: [describe pipeline stages and transformations]
Automated lineage documentation is invaluable for regulatory compliance, debugging, and onboarding new engineers to complex pipelines.
Data Governance & Quality
Prompt 11: Data Quality Rules Catalog
For the following dataset and its intended business use, generate a comprehensive data quality rules catalog. Include completeness, validity, consistency, timeliness, and uniqueness rules. For each rule, specify the check logic, severity level (warning vs. blocking), and recommended remediation action: [describe dataset and use case]
A well-structured quality rules catalog is the foundation of any automated data quality framework and reduces the time to implement monitoring by providing ready-to-implement specifications.
Prompt 12: Master Data Management Strategy
Design a master data management (MDM) strategy for [entity name, e.g., Customer or Product] across the following source systems: [list systems and their data characteristics]. Include golden record definition, survivorship rules, match/merge logic, and a stewardship workflow.
MDM strategy documents require deep domain and technical knowledge simultaneously; this prompt helps you produce a comprehensive first draft that domain experts can then refine.
Prompt 13: Data Classification Framework
Create a data classification framework for an enterprise with the following regulatory obligations and data types: [list regulations such as GDPR, HIPAA, CCPA and data types]. Define classification tiers, labeling conventions, handling requirements per tier, and access control policies.
A clear classification framework is a prerequisite for meaningful data governance and dramatically reduces the effort required for subsequent security and compliance work.
Prompt 14: Data Ownership RACI Matrix
Generate a RACI matrix for data governance activities across the following roles and data domains in our organization: [list roles and data domains]. Activities should include data definition, quality monitoring, access approval, incident response, and retention enforcement.
A RACI matrix removes ambiguity about accountability, which is the single most common reason data governance programs stall or fail.
Prompt 15: Regulatory Compliance Gap Analysis
Conduct a gap analysis of the following data architecture against [regulation, e.g., GDPR Article 17 Right to Erasure]. Identify gaps, assess their risk level, and propose remediation steps including estimated implementation complexity: [describe current architecture and data flows]
Regulatory gap analyses are time-intensive when done manually; this prompt produces a structured output that legal and engineering teams can work from immediately.
Cloud Data Platform Strategy
Prompt 16: Cloud Data Platform Evaluation
Evaluate the following three cloud data platform options — [e.g., Snowflake, Databricks, BigQuery] — against our requirements: [list workload types, scale, latency needs, team skills, and budget constraints]. Provide a scoring matrix and a final recommendation with rationale.
A structured platform evaluation reduces the risk of vendor lock-in decisions being made on marketing materials rather than engineering criteria.
Prompt 17: Lakehouse Architecture Design
Design a lakehouse architecture for the following organization: [describe industry, data sources, consumer types — BI, ML, operational — and scale]. Include storage layer design, table format choice (Delta, Iceberg, or Hudi), catalog strategy, compute separation approach, and governance integration points.
Lakehouse architecture involves numerous interrelated decisions that are easy to get wrong in isolation; this prompt enforces holistic thinking across all layers.
Prompt 18: Multi-Cloud Data Strategy
Our organization needs a multi-cloud data strategy spanning [list cloud providers and their current workloads]. Design an architecture that minimizes data egress costs, avoids vendor lock-in, ensures consistent governance, and supports federated query across clouds.
Multi-cloud strategies require careful cost modeling and architectural discipline; this prompt surfaces the trade-offs that are often discovered too late in implementation.
Prompt 19: Cloud Cost Optimization Audit
Audit the following cloud data platform configuration for cost optimization opportunities: [describe current setup including storage tiers, compute configurations, query patterns, and monthly spend breakdown]. Prioritize recommendations by potential savings and implementation effort.
Cloud data costs can scale unpredictably with growth; a systematic audit prompt produces an actionable savings plan rather than generic advice.
Prompt 20: DataOps Implementation Roadmap
Create a DataOps implementation roadmap for an organization currently at [describe current maturity: ad-hoc pipelines, manual deployments, limited monitoring]. Define maturity stages, key capabilities to build at each stage, tooling recommendations, and a 12-month milestone plan.
DataOps transformation requires a phased approach that balances quick wins with long-term structural improvements; this prompt produces a roadmap that leadership can approve and engineering can execute.
Stakeholder Communication & Documentation
Prompt 21: Executive Architecture Summary
Translate the following technical data architecture design into a one-page executive summary suitable for a C-suite audience. Emphasize business value, risk reduction, and strategic alignment. Avoid jargon and use analogies where appropriate: [paste technical design]
The ability to translate technical architecture into business language is often the difference between getting a project funded and watching it die in committee.
Prompt 22: Architecture Decision Record
Write a formal Architecture Decision Record (ADR) for the following decision: [describe the decision, context, and options considered]. Include sections for Context, Decision, Consequences (positive and negative), Alternatives Considered, and Review Date.
ADRs create an institutional memory that prevents the same architectural debates from being relitigated every time team membership changes.
Prompt 23: Data Dictionary Generation
Generate a comprehensive data dictionary for the following table or dataset. For each field, include: business name, technical name, data type, nullable flag, business definition, example values, valid value ranges or enumerations, and the source system of record: [paste schema or field list]
A well-written data dictionary dramatically reduces the time analysts and engineers spend tracking down field definitions and reduces misinterpretation of data.
Prompt 24: Stakeholder Requirements Workshop Agenda
Design a structured workshop agenda for gathering data architecture requirements from the following stakeholder groups: [list groups and their known concerns]. Include icebreaker activities, structured elicitation exercises, prioritization activities, and a decision-capture mechanism. Duration: [X hours].
A well-designed requirements workshop agenda ensures all stakeholder voices are heard and produces structured outputs rather than unactionable meeting notes.
Prompt 25: Technical Proposal for Non-Technical Audience
Rewrite the following technical data architecture proposal for a non-technical business audience. Replace technical terms with plain-language equivalents, add a business impact section, and include a simple visual description of the proposed architecture using only text: [paste proposal]
Bridging the communication gap between data engineering and business stakeholders is a core competency for senior data architects and this prompt makes that translation systematic.
Performance Optimization
Prompt 26: Query Performance Diagnostic
Act as a database performance engineer. Analyze the following slow query and its execution plan. Identify the root causes of poor performance and provide specific, prioritized optimization recommendations including index changes, query rewrites, statistics updates, and schema modifications: [paste query and execution plan]
Query optimization is a high-skill, time-intensive task; this prompt accelerates diagnosis and produces a prioritized action list that engineers can implement immediately.
Prompt 27: Partitioning Strategy Design
Design an optimal partitioning strategy for the following table given these query patterns and data volume: [describe table, query patterns, row counts, and growth rate]. Compare partition key options, estimate the performance impact of each, and recommend a strategy with migration steps.
Partitioning decisions have enormous performance implications that are difficult to change after the fact; a structured analysis prevents costly rearchitecting later.
Prompt 28: Indexing Strategy Audit
Audit the following table's current index configuration against these common query patterns. Identify redundant indexes, missing indexes, and index bloat issues. Provide a recommended final index set with justification for each: [paste table definition, existing indexes, and top 10 queries by frequency]
Over-indexing and under-indexing are both common and costly; this audit prompt produces a minimal, well-justified index set rather than ad-hoc additions over time.
Prompt 29: Caching Architecture Design
Design a caching architecture for the following data platform workload: [describe query patterns, data freshness requirements, cache invalidation triggers, and scale]. Include cache layer placement, technology selection, TTL strategy, cache warming approach, and monitoring plan.
Effective caching can reduce compute costs and latency by orders of magnitude, but poorly designed caching introduces staleness and consistency bugs; this prompt ensures all dimensions are covered.
Prompt 30: Capacity Planning Model
Build a capacity planning model for the following data platform over a 24-month horizon: [describe current storage volumes, growth rates, query volumes, and infrastructure configuration]. Project storage, compute, and network requirements by quarter, and identify the top scaling bottlenecks to address proactively.
Capacity planning prevents the reactive, expensive infrastructure expansions that occur when growth outpaces planning, and a structured model enables proactive budget requests.
Career Development
Prompt 31: Skills Gap Analysis
I am a data architect with the following skills and experience: [describe background]. I want to move toward [target role or specialization, e.g., Principal Data Architect, Chief Data Officer, ML Platform Engineer]. Conduct a skills gap analysis and create a 12-month learning roadmap with specific resources, projects, and milestones.
A personalized skills gap analysis gives your career development a strategic direction rather than leaving it to chance or generic advice.
Prompt 32: Portfolio Project Design
Design a portfolio project that demonstrates advanced data architecture skills to potential employers or leadership. The project should showcase [target skills: e.g., lakehouse design, real-time streaming, data governance]. Include a project brief, technical scope, deliverables list, and GitHub repository structure.
A well-scoped portfolio project signals seniority to hiring managers and promotion committees more effectively than a resume bullet point.
Prompt 33: Technical Interview Preparation
I have a technical interview for a [specific role] at a [industry] company. Generate 20 advanced data architecture interview questions I am likely to face, along with guidance on what a strong answer should cover for each question. Focus on system design, trade-off analysis, and real-world scenario questions.
Preparing for the specific types of questions that senior architecture roles require — rather than generic data engineering questions — dramatically improves interview performance.
Prompt 34: Conference Talk Proposal
Help me write a compelling conference talk proposal for [conference name] based on the following project or experience I want to share: [describe project or insight]. Include a title, abstract (300 words), key takeaways for the audience, and a high-level outline of the session structure.
Public speaking at industry conferences is one of the highest-leverage career moves for data architects, building reputation and network simultaneously; a strong proposal is the first gate.
Prompt 35: Personal Brand Content Strategy
I am a data architect specializing in [your specialization] and I want to build a professional presence on [platform, e.g., LinkedIn, Dev.to, Substack]. Create a 90-day content strategy including topic pillars, a content calendar with 12 specific post ideas, and guidance on the format and tone that resonates with my target audience of [describe audience].
A consistent personal brand built around genuine expertise creates inbound career opportunities — job offers, consulting inquiries, speaking invitations — that would otherwise never materialize.
Want all 35 prompts in a convenient, copy-paste format? Get the complete AI Prompt Toolkit for this profession →
Top comments (0)