3. Data, Science & AI

Adds and configures new or custom AI models (including self-hosted and non-standard providers) to opencode.json via research and verification.

Provides principles for context-efficient design of LLM agents and multi-agent systems.

Generates a context document by analyzing project structure to reduce token usage and enhance AI agent performance in codebase understanding.

Creates optimized relational database schemas with proper relationships and normalization for efficient data storage.

Tracks data lineage from source to consumption for auditing, debugging, compliance, and dependency analysis in data workflows.

Creates optimized relational database schemas with normalized tables, relationships, and appropriate data types for efficient data storage and retrieval.

Merges and reconciles data from multiple sources, resolving conflicts and ensuring quality for unified data warehouses and master data systems.

Designs scalable database schemas for SQL and NoSQL databases, covering normalization, indexing, partitioning, and migration strategies to support growth from prototype to production.

Architects scalable data infrastructure for pipelines, storage, and distributed systems to handle high throughput and production demands.

Analyzes sentiment in financial news and social media (Twitter/Reddit) to track market trends and public opinion.

Systematically evaluates LLM and AI product performance through test case design, accuracy measurement, regression tracking, and continuous benchmarking pipelines.

Provides real-time and historical market data for A-shares, US stocks, and Hong Kong stocks, including K-lines, indices, sector rankings, and capital flow for basic analysis.

Provides US-China macroeconomic data (GDP, CPI, PMI), global risk indicators (VIX, DXY), and economic calendar events for investment research.

Guides reliable autonomous AI agent implementation through goal decomposition, reflection patterns, and error mitigation to prevent compounding failure rates.

Analyzes contracts using NLP to extract key terms, identify risks, and score potential issues for compliance and business decisions.

Provides mathematical modeling tools for pricing options, calculating Greeks, and implementing exotic derivatives using Black-Scholes, Monte Carlo, and QuantLib.

Provides real-time hand gesture recognition and touchless interface design using computer vision and MediaPipe for sign language interpretation and UI interactions.

Provides patterns for building robust, reproducible genomics analysis pipelines including NGS processing, variant calling, and RNA-seq.

Provides engineering tools for power systems analysis including grid modeling, power flow, energy storage dispatch, and electricity market economics.

Builds and maintains virtual replicas of physical systems for real-time monitoring, predictive analytics, and optimization through data synchronization and AI-driven modeling.

Evaluates LLM agent performance via behavioral testing, capability assessment, and reliability metrics to measure real-world effectiveness.

Integrates local and cloud LLMs into Unreal Engine for dynamic NPC dialogue and intelligent behaviors using Blueprints and C++.

Provides multi-sensor data fusion patterns including Kalman filters, complementary filters, and factor graph optimization for state estimation in robotics and autonomous systems.

Creates AI-native brand systems to ensure consistency in AI-generated content via brand guideline training and reusable prompt libraries.