Top AI Prompts for Data Science Success

Identify 5 promising feature engineering strategies for a customer churn dataset, based on insights from the ‘Churn Analysis 2024’ report.

Use Case: Speeds up feature selection by leveraging past project learnings.

ClickUp Brain Behaviour: Extracts key techniques and patterns from linked documents to suggest effective feature approaches.

What are the current best practices for hyperparameter tuning in gradient boosting models in finance?

Use Case: Guides model optimization with up-to-date domain-specific methods.

ClickUp Brain Behaviour: Synthesizes recommendations from internal research and can incorporate external publications if available.

Draft a project brief for building an explainable AI model for credit scoring, referencing ‘Explainability Guidelines’ and prior project notes.

Use Case: Aligns data science and compliance teams with a clear, shared objective.

ClickUp Brain Behaviour: Compiles relevant document excerpts into a structured, actionable brief.

Summarize performance benchmarks between XGBoost and LightGBM on fraud detection datasets using the ‘Model Comparison Q1’ report.

Use Case: Facilitates quick comparative insights without manual data review.

ClickUp Brain Behaviour: Extracts and condenses tabular and textual data into a concise comparison summary.

List top data preprocessing techniques for handling missing values in healthcare datasets, referencing R&D notes and best practice documents.

Use Case: Supports selection of robust cleaning methods for sensitive data.

ClickUp Brain Behavior: Scans internal documents to identify commonly recommended approaches and their effectiveness.

From the ‘Model Validation Checklist’ doc, generate a detailed QA checklist for data pipeline testing.

Use Case: Simplifies validation planning with a ready-to-use task list.

ClickUp Brain Behavior: Identifies key validation steps and formats them into a structured checklist within a task or document.

Summarize 3 emerging trends in natural language processing for customer support from recent research and internal reviews.

Use Case: Keeps NLP model development aligned with cutting-edge techniques.

ClickUp Brain Behavior: Extracts recurring themes and insights from linked research documents.

From the ‘User Feedback Q2’ doc, summarize main user preferences for dashboard visualizations.

Use Case: Helps design teams tailor analytics interfaces to user needs.

ClickUp Brain Behavior: Analyzes survey data to highlight common visualization preferences and pain points.

Write concise and engaging tooltip text for a data quality alert feature, using the style guide in ‘UXTone.pdf’.

Use Case: Accelerates UI copywriting while maintaining brand voice.

ClickUp Brain Behavior: References tone guidelines to generate multiple copy options for interface elements.

Summarize key updates in GDPR data handling regulations for AI models and their implications on feature selection.

Use Case: Ensures compliance considerations are integrated into model development.

ClickUp Brain Behavior: Condenses regulatory documents and highlights relevant compliance points; can access public updates if linked.

Generate guidelines for data anonymization techniques compliant with HIPAA, referencing internal policy documents.

Use Case: Supports secure data processing in healthcare projects.

ClickUp Brain Behavior: Extracts procedural rules and best practices to form a compliance checklist.

Create a checklist for model deployment readiness using the ‘Deployment Standards’ folder and recent project retrospectives.

Use Case: Helps teams ensure all criteria are met before production rollout.

ClickUp Brain Behavior: Identifies deployment requirements and organizes them into actionable tasks grouped by category.

Compare data augmentation techniques used in image classification projects across recent internal studies.

Use Case: Informs strategy for improving model robustness.

ClickUp Brain Behavior: Summarizes documented methods and their reported effectiveness in a clear format.

What are the latest trends in automated machine learning tools since 2023?

Use Case: Guides R&D teams on adopting efficient model building platforms.

ClickUp Brain Behavior: Synthesizes trends from internal analyses, vendor reports, and uploaded research papers.

Summarize key data quality issues reported in Southeast Asia market datasets, focusing on missingness, inconsistencies, and outliers.

Use Case: Drives targeted data cleaning efforts for regional projects.

ClickUp Brain Behavior: Extracts and prioritizes user-reported data problems from feedback forms, surveys, and issue trackers.

Define your data science scenario clearly

Vague prompts yield broad results. Specify details like dataset type (e.g., “time series sales data” or “customer churn records”), analysis goals (e.g., “forecasting demand” or “segmenting users”), or business context (e.g., “retail market in Q4”).

Example: “Propose feature engineering techniques for predicting churn in telecom customer data.”

Use comparative prompts to reveal distinctions

AI excels at contrasting alternatives. Frame prompts as comparisons to assess model performances, evaluate algorithms, or benchmark datasets.

Example: “Compare accuracy and training time between random forest and gradient boosting on fraud detection data.”

Focus prompts on specific data tasks

Treat your prompt as a clear data science objective. Instead of vague requests like “Generate models,” specify the task:

Example: “Design a clustering approach to identify customer segments in e-commerce transaction data.”

Specify desired output formats

Need a summary table, code snippet, visualization plan, or step-by-step methodology? Indicate it explicitly. AI delivers better when output expectations are clear.

Example: “Provide a bullet list of key preprocessing steps for handling missing values in healthcare datasets.”