3. Data, Science & AI

Enables analytics engineering with dbt for organizing data models, testing, documentation, and incremental data transformations.

Optimizes Apache Spark job performance through partitioning, caching, shuffle tuning, and memory configuration for faster data processing pipelines.

Builds robust backtesting systems for trading strategies, handling look-ahead bias, survivorship bias, and transaction costs to validate algorithmic models.

Provides best practices for building production-ready Apache Airflow DAGs for data pipeline orchestration and workflow management.

Implements data quality validation using Great Expectations, dbt, and data contracts for robust data pipelines.

Provides foundational knowledge on context management in AI agent systems for architecture design, debugging, and optimization.

Provides A-share quantitative trading analysis and backtesting using Tushare Pro data and Backtrader framework.

Configures Sentry for real-time monitoring of AI agent interactions and LLM API calls across supported frameworks like OpenAI and LangChain.

Orchestrates multi-agent workflows by delegating tasks to specialist sub-agents while maintaining context and preserving implementation autonomy.

Executes PostgreSQL database queries and meta-commands directly via command line for data retrieval and management.

Fetches real-time and historical market data, financial news, and fundamental data to support trading and investment decisions.

Generates institutional-grade investment reports with AI-driven technical analysis, charts, and recommendations for stocks and ETFs.

Expert in parsing, transforming, and validating data to ensure accuracy and usability for analysis or integration.

Provides fundamental tools for astronomical data processing including FITS I/O, coordinate transformations, and time handling using Astropy.

Manages scientific Python environments with unified conda-forge and PyPI support, enabling reproducible workflows for complex computational packages.

Enables efficient handling of labeled multidimensional scientific datasets, including climate, satellite, and oceanographic data, with support for NetCDF/HDF5 and Dask.

Builds and optimizes PySpark data pipelines on Databricks for scalable data processing and analytics workloads.

Assesses codebase maturity for autonomous AI development by evaluating eight technical pillars and five maturity levels to identify gaps.

Detects deepfakes using multimodal forensic analysis with PRNU, IGH, DQ, and LLM-augmented techniques for media authentication.

Provides PostgreSQL performance optimization and best practices including query tuning, connection pooling, and schema design.

Enables querying, updating, and deleting MongoDB collections and documents in IN7DB and AgentDB for data verification and debugging.

Enables users to request second opinions from other AI models via Gemini CLI or orcx interface.

Initializes context management for AI agents in a project, enabling structured state handling and interaction workflows.

Reduces context consumption in AI systems by loading agent definitions only when required, optimizing LLM resource usage.