3. Data, Science & AI

Provides best practices for database design, Row Level Security (RLS), and performance optimization in Supabase for ERP applications.

Provides precise legal answers by performing RAG-based searches on the Russian Constitution, citing exact article references.

Ensures sufficient context is present before AI actions, preventing incomplete or erroneous responses in agent systems.

Standardizes drillhole data handling, section interpolation, and three-tier validation for geological analysis workflows.

Converts markdown-based Product Requirement Documents (PRDs) into structured JSON format for automated data processing and system integration.

Creates synthetic datasets via NeMo Data Designer configuration and preview execution for AI model training and data augmentation.

Dispatches independent tasks to parallel agents without shared state or sequential dependencies.

Rotates LLM context to prevent pollution, persisting state in files and git for autonomous development.

Automates creation of Claude Code subagents using tmux and /agents TUI for efficient AI agent management and deployment.

Searches semantic memory for relevant failures and patterns using meaning-based retrieval.

Expert guidance for AI SDK Core features: text generation, structured data, tool calling, embeddings, and MCP integration for LLM applications.

Expert guidance for building AI agents with ToolLoopAgent (AI SDK v6+), covering configuration and workflow patterns.

Performs pre-execution risk assessment to prevent AI hallucinations and maintain architectural integrity in model deployments.

Converts HVDC JSON data to RDF Turtle format and validates RDF conformance.

Enhances text translation quality in images while preserving layout for charts, infographics, and PDFs through OCR and multilingual re-rendering.

Validates code and provides physics reasoning for 2D doublet models using streamline methods and transport calculations in computational physics.

Provides standardized guidelines for developing efficient AI agent skills with optimized token usage to reduce costs and enhance performance.

AI orchestrator leveraging semantic memory graphs to autonomously explore, plan, and build solutions through self-learning.

Orchestrates AI agents to initiate and manage end-to-end development workflows, enhancing coding efficiency.

Displays real-time health metrics for learning systems, including memory counts, edge counts, effectiveness, and feedback loop status.

Provides accurate PostgreSQL documentation guidance for SQL syntax, configuration, performance tuning, and database management tasks.

Validates code and provides physics reasoning for a 0-D CO2 lumped-parameter model in geothermal reservoir simulations, including ODE solvers and emissions calculations.

Optimizes database performance through query tuning, indexing, caching, and cloud strategies to resolve scalability and speed issues.

Creates and documents AI agent skills following the Agent Skills specification for specialized agent capabilities.