3. Data, Science & AI

Automatically analyzes user feedback and conversation patterns to update project skills with learned preferences for improved future interactions.

Python framework for developing multi-agent AI systems using OpenAI's agent capabilities to automate complex task workflows.

Programmatically creates, manages, and executes Databricks notebooks for data pipeline automation, ETL workflows, and data processing tasks.

Enables complex reasoning through structured multi-LLM workflows, supporting task decomposition, perspective integration, and iterative refinement for high-quality outputs.

Executes SQL queries against Databricks, including Unity Catalog tables, with result formatting and error handling for data analytics and exploration.

Profiles datasets (CSV, JSON, Parquet) to analyze schema, quality metrics, distributions, and uniqueness for data analysis workflows.

Manages Unity Catalog metadata tags for data governance, classification, and compliance workflows, including sensitivity tagging (PII, PHI) and data quality attributes.

Provides competitive intelligence data analysis methods, including comparative frameworks, chart selection, and visualization standards.

Comprehensive guide for building AI agents using Google's Agent Development Kit (ADK), covering agent types, tool integration, and multi-agent workflows for LLM-powered applications.

Configures AI agent profiles, capabilities, tools, and complex workflows for runtime orchestration.

Optimizes user input context for better AI model performance using context engineering techniques, specifically for Claude when requested or with the --foryou flag.

Generates AntV Infographic DSL syntax for creating data visualizations, handling template selection, data structuring, and theming.

Provides advanced prompt engineering techniques for Claude models, including system prompts, chain of thought, and multi-agent workflows.

Provides best practices for relational database schema design, including ERD creation, DDL scripting, and indexing strategies.

Systematically evaluates LLM application performance using RAGAS metrics for retrieval and generation quality.

Extracts structured data from neurosurgical research studies for analysis and database integration.

Performs network meta-analysis to compare multiple treatments by synthesizing direct and indirect evidence from clinical trials.

Meta-skill for creating, refining, and managing AI agent skills, enabling self-improvement and knowledge capture through automated skill development.

Coordinates multiple AI agents to collaborate on complex tasks through efficient orchestration of their execution flows.

Guide for building AI agents with Google ADK, covering multi-agent pipelines, agent patterns, and debugging data flow issues.

Executes statistical synthesis of multiple studies, including data pooling, heterogeneity assessment, and forest/funnel plot generation.

Monitors protocol KPIs and TVL trends, generates executive summaries, and flags at-risk vaults for internal operations teams.

Guides PostgreSQL query optimization with EXPLAIN analysis, index recommendations, and solutions for common performance issues.

Provides Trial Sequential Analysis (TSA) for meta-analyses with information size calculations to ensure statistical validity.