3. Data, Science & AI

Verifies that software implementations adhere to mathematical and physical principles.

Creates Jupyter Notebooks (.ipynb) containing explanatory content for data science features and themes.

Automates sequential execution of project Jupyter notebooks to validate error-free operation and ensure data pipeline reliability.

Estimates time and LLM token consumption from work plans to evaluate cost-performance for AI operations.

Provides extraction logic and rendering API for multi-dimensional field data (e.g., Field4D, Matrix) while maintaining physical consistency.

Analyzes technical and physical differences between similar signal processing methods in a library, generating comparative notebook documentation.

Comprehensive agent skills for context engineering, multi-agent architectures, and debugging in production AI systems.

Proposes the optimal large language model (LLM) for specific tasks including coding, research, and refactoring.

Consults multiple LLM providers to generate design options, risk checks, and alternative solutions before implementing code changes.

Manages and optimizes LLM usage quotas to prevent overages and ensure continuous access.

Generates reusable agent skills from conversation history and work content for AI assistant customization.

Enables intelligent planning and coordination of multiple agents for complex multi-step tasks through decomposition, dependency mapping, and dynamic execution strategies.

Comprehensive toolkit for web crawling, structured data extraction, and automated data pipeline building without LLM dependencies.

Calculates asset price deviation from long-term exponential trend to identify historical extremes and assess market conditions via optional macro factor analysis.

Analyzes silver price movements using palladium trends and industrial/risk sentiment indicators to identify confirmed market directions.

Calculates non-linear impact of trading costs (commissions/spreads) on risk-reward ratios using cost density metric to determine optimal stop-loss levels and profit thresholds.

PostgreSQL database specialist for query optimization, schema design, security, and performance with Supabase best practices.

Tracks and visualizes cumulative returns of multiple stocks/indices against S&P 500 benchmark, including top N rankings for comparative analysis.

Safely handles database schema changes with backup, sequential execution, and rollback capabilities for applications like Budget Buddy.

Debugs and tests Budget Buddy's fuzzy matching transaction classification system using 0.85 similarity threshold for merchant matching and smart batch updates.

Provides expert guidance on ClickHouse database patterns, query optimization, and data engineering best practices for high-performance analytics.

Provides agent-first workflow and design framework for building X07 AI programs with the released toolchain, executed via `x07 run`.

Automates web data extraction and API data retrieval via customizable scraping scripts.

Provides historical data-driven entry signal patterns for trading decisions, based on past success rates to inform position openings.