3. Data, Science & AI

Optimizes LLM performance and output reliability through advanced prompt engineering patterns for production deployment.

Assists in database design principles including schema creation, indexing strategies, ORM selection, and serverless database options like Supabase.

Removes context-specific elements from AI-generated artifacts to enhance reusability and general applicability.

Builds knowledge models from scratch using distributed hierarchical architecture, centrally defining metamodels and generating instance folders.

Sets up hierarchical Intent Layer (AGENTS.md) for codebases to enable AI agents to understand project structure and context.

Enables interactive data visualization, statistical analysis, and data profiling for CSV files using Plotly.

Enables parallel execution of specialized AI agents for multi-perspective analysis and task decomposition in complex problem-solving scenarios.

Optimizes content for generative AI engines like ChatGPT, Claude, and Perplexity to enhance visibility and relevance in AI-driven search results.

Manages local knowledge bases, embeddings, and enables RAG for efficient information retrieval and AI-driven responses.

Clarifies ambiguous user requests through structured prompt interpretation to enable precise AI responses.

Integrates MiniMax MCP server with Claude for token-efficient AI delegation, enabling web search, image analysis, and collaboration via terminal/desktop scripts and slash commands.

Transforms data into compelling narratives for stakeholder presentations, data reports, and executive summaries using visualization and persuasive structure.

Enables token-efficient AI delegation between Claude and MiniMax MCP server for web search, image analysis, and terminal/desktop collaboration via scripts and slash commands.

Enables data visualization and DataFrame inspection using matplotlib/seaborn, supporting file and Marimo notebook data sources.

Generates DuckDB SQL queries for data analysis on CSV, Parquet, and .ddb files, enabling efficient data exploration and transformation.

Provides AI-driven quality assessment of game screenshots and sprites using GLM-4.6v model for technical evaluation.

Autonomous AI agent framework for iterative feature implementation, orchestrating AI tools to complete user stories through quality gates and tracking progress until requirements are met.

Analyzes business sales and revenue data from CSV files to identify weaknesses, generate statistical insights, and provide strategic improvement recommendations for performance reports.

Monitors Twitter/X watchlists, uses AI to filter valuable content, and generates magazine-style Markdown reports for tech insights and AI trends.

Generates daily comprehensive reports for personal Twitter/X account analytics, including engagement metrics and content performance.

Orchestrates complex multi-step tasks by delegating to parallel subagent teams, preserving context and avoiding auto-compact in LLM workflows.

Provides reference for Salesforce object schemas, field types, and relationships to enable accurate SOQL query development.

Open-source tool for web scraping and data processing with intelligent crawling, screenshots, and PDF export capabilities.

Executes SOQL queries and manages Salesforce authentication via CLI, enabling efficient data retrieval and connection handling.