3. Data, Science & AI

Generates structured plot workflows for long stories by orchestrating high-level and detailed plot points via modular AI agents.

Analyzes multi-round evaluation scoring data to compute metrics and assign rating grades (S/A/B), enabling trend analysis and performance assessment.

Analyzes multi-round evaluation scoring data, calculates metrics, and determines rating levels (S/A/B) for trend analysis and performance assessment.

Formats agent outputs into structured data for reports, supporting multiple formats like JSON, XML, and Markdown.

AI-driven evaluation of novel manuscripts scoring market potential, innovation, and content quality for initial screening and multi-dimensional analysis.

Analyzes user-uploaded file references to extract key information and provide application recommendations, including @symbol and natural language parsing for vertical short drama planning.

Analyzes user-uploaded file references to extract key information and provide application suggestions, including parsing @-mentions and natural language references.

Designs and implements multi-agent AI systems using coordination patterns like orchestration, delegation, and parallel processing for complex task execution.

Provides best practices and guidance for authoring Apache Airflow Directed Acyclic Graphs (DAGs) to define and manage data workflows and pipelines.

Queries data warehouse to answer business questions via SQL, including metrics, trends, and data lookups.

Quickly checks data freshness, last update times, and staleness for database tables and datasets.

Traces upstream data lineage to identify data sources, dependencies, and origins for tables and datasets.

Traces downstream data dependencies and assesses impact of changes to data sources, tables, or pipelines to mitigate risk.

Provides statistical analysis, data quality insights, and structural overview for a specific database table upon request.

Creates a markdown file with table metadata for instant data warehouse schema lookups, enabling efficient project-wide data discovery.

Manages Apache Airflow operations for orchestrating data pipelines, including DAG execution, monitoring, and debugging.

Engineers effective prompts for Claude using systematic methodology, balancing simplicity, cost, and effectiveness through progressive disclosure and empirical validation.

Verifies structure, build, training success, and metric logging of a machine learning training pipeline for validation and debugging.

Verifies structure, build, and correct execution of machine learning inference examples to ensure proper implementation and debugging.

Analyzes machine learning model architectures from research papers and code to facilitate implementation understanding.

Processes and validates datasets for machine learning training, ensuring data quality and readiness in data pipelines.

Extracts and documents machine learning model hyperparameters from research papers to streamline training configuration setup.

Executes machine learning model training using optimization algorithms on datasets, managing training loop operations.

Validates agent hierarchy and escalation paths in AI systems through delegation pattern testing after structural changes.