3. Data, Science & AI

Generates optimal pacing and fueling strategies for athletic races by analyzing current fitness levels and course-specific constraints.

Enables creation, editing, and analysis of spreadsheets with formula handling, data visualization, and support for multiple spreadsheet formats.

Creates custom AI tools for Vercel AI SDK agents, enabling specialized function integration into agent workflows.

Analyzes MLflow experiment traces to debug, investigate, and resolve issues in machine learning workflows.

Analyzes MLflow sessions to debug multi-turn chat conversations, review history, and identify issues in LLM interactions.

Manages database schema migrations for Neon PostgreSQL using Drizzle ORM, enabling version-controlled database evolution and troubleshooting.

Provides post-run analysis with effort assessment, recovery advice, and next-session guidance based on completed training sessions.

Generates personalized 14–21 day running plans using user onboarding data and recent performance metrics.

Emulates Aristotle system for autonomous theorem proving in Lean 4, accepting theorem statements and optional natural language proof sketches.

Guides developers in building LLM applications with LangChain, covering chains, prompts, memory, and retrievers for efficient AI workflows.

Guides new users through MLflow setup by identifying their use case (GenAI or traditional ML) and providing tailored quickstart tutorials and integration steps.

Customizes data plot aesthetics (colors, fonts, titles, labels) via natural language for publication-ready visualizations.

Simulates real user interactions with UIs using LLMs to test and improve user experience through structured protocols and best practices.

Searches and retrieves historical data visualizations for reuse, analysis, or export in data science workflows.

Provides strategies for managing LLM context windows to handle token limits and maintain context integrity in long conversations or documents.

Guides autonomous agent development for Claude Code plugins, covering structure, system prompts, tools, and best practices.

Guides database schema design, SQL optimization, and data modeling best practices for efficient data management and security implementation.

Creates markdown reports with embedded data visualizations for analysis documentation, sharing, and result export.

Guides users through conversational requirement gathering to produce Dify-compatible workflow DSL YAML for immediate AI application deployment.

Provides Supabase-curated best practices for optimizing PostgreSQL performance, including query tuning, schema design, and configuration.

Guides creation, configuration, and utilization of Claude Code subagents and the Task tool for specialized AI agent workflows.

Generates actionable sales pipeline health reports by querying HubSpot data through Metabase, covering SMB, Mid-market, Enterprise, and Add-on segments for leadership meetings.

Builds resilient pipelines for incremental data ingestion from paginated APIs, handling rate limits, duplicates, and historical backfills without re-downloading.

Extracts hidden patterns and actionable insights from stored data using AI-driven analysis.