3. Data, Science & AI

Handles session termination and context updates for AI models during handoffs, preserving state without judgment for efficient orchestrator transitions.

Fetches oil/gas trade flow data including trade queries, aggregations, vessel positions, and company fleet information via Kpler API.

Fetches financial data from LSEG Workspace using Eikon Data API, handling cURL requests, field discovery, and time series retrieval.

Guides AI agents through behavior data processing workflows for Sun Lab VR sessions using MCP server, with documentation of log formats and hardware data structures.

Enables coordination between multiple AI agents for collaborative coding tasks, facilitating communication and task delegation across Claude Code sessions.

Standardizes bioinformatics project directories with separation of raw data, code, analysis, results, and temporary files, integrating cloud and Git workflows.

Automates UI visual verification using computer vision AI to validate design compliance and detect visual discrepancies.

Streamlines AI agent application development with code generation, model comparison, and evaluation setup for efficient workflow creation.

Enriches retail location data by researching web sources to populate missing fields in databases for shopping centres, retail parks, and outlet centres.

Generates Pareto frontier visualizations for multi-objective optimization problems, showing non-dominated solutions to aid trade-off decision-making.

Objective weighting method using information entropy to assign higher weights to indicators with greater variability, reducing subjectivity in decision analysis.

Simulates pattern formation and spatial dynamics using reaction-diffusion PDEs for ecological, biological, and environmental applications.

Reviews database migrations and data model changes for safety, data constraints, transaction boundaries, and referential integrity.

Provides ensemble regression models using Random Forest and XGBoost for accurate predictions and feature analysis in data-driven modeling.

Comprehensive framework for multi-objective optimization using NSGA-II, MOEA/D, and TOPSIS, generating Pareto frontiers and quality metrics for engineering and policy decisions.

Performs Principal Component Analysis for dimensionality reduction, feature extraction, and visualization of high-dimensional datasets.

Performs classical time series forecasting with ARIMA/SARIMA models, capturing trends, seasonality, and confidence intervals for temporal prediction.

Provides implementation and analysis of Lotka-Volterra equations for modeling predator-prey dynamics and population interactions in ecological and competitive systems.

Ranks alternatives using geometric distance to ideal solutions in multi-criteria decision analysis.

Unified framework for mechanistic modeling with ODEs/PDEs, including parameter estimation, stability analysis, and visualization for scientific workflows.

Automates data cleaning for MCM/ICM competitions, handling missing values, outliers, normalization, and generating statistical summaries.

Validates database migration PRs for data integrity, ID mapping accuracy, and rollback safety to prevent production data issues.

Calculates subjective weights via pairwise comparison matrices using eigenvalue method and consistency ratio check.

Provides a Genetic Algorithm implementation for solving combinatorial optimization problems like TSP, scheduling, and assignment using evolutionary principles.