3. Data, Science & AI

Provides tools for detrending economic time series using HP filter, log transformations, and business cycle correlation analysis.

Provides comprehensive SQL patterns for database operations and migrations across PostgreSQL, MySQL, and SQLite.

Analyzes user retention and churn using survival analysis, cohort analysis, and machine learning to predict churn risk and optimize retention strategies.

Enables efficient mathematical proof writing in Lean 4 with error resolution, mathlib lemma search, and LSP integration for formal verification.

Generates and optimizes SQL queries for efficient data retrieval and analysis, enhancing data processing workflows.

Searches a bundled flights dataset for available routes and dates, providing flight options or checking route/date availability.

Repairs infeasible flexible job shop scheduling plans to be downtime and precedence feasible without worsening the policy budget.

Searches a bundled dataset for accommodations in a specified city, enabling recommendations and filtering for travel itineraries.

Provides molecular data processing including SMILES parsing, descriptor calculation, and substructure search for cheminformatics applications.

Builds and trains ML models for classification, regression, and clustering using scikit-learn, PyTorch, and TensorFlow.

Validates datasets, model checkpoints, and training pipelines for ML projects to ensure data quality and pipeline integrity.

Harmonizes clinical laboratory data across sources by converting units, standardizing numeric formats, and resolving quality issues for healthcare analytics and research.

Performs nonlinear least squares parameter estimation for experimental data using SciPy's curve_fit function.

Detects flood events by comparing water levels to thresholds, enabling occurrence determination, day counting, and severity classification.

Estimates driving/taxi duration, distance, and cost between cities using a pre-bundled distance matrix CSV.

Implements mHC algorithm for stabilizing deep neural network training using manifold constraints and Sinkhorn-Knopp for doubly stochastic matrices.

Enables high-performance numerical computing and machine learning with JAX, featuring automatic differentiation, JIT compilation, and array operations for scientific and ML applications.

Detects exoplanet transits in astronomical light curves using the Transit Least Squares algorithm.

Converts PDFs to Markdown while preserving LaTeX formulas, equations, and document structure for academic and technical documents.

Programmatically process PDFs: extract text/tables, create, merge, split, and fill forms for document analysis and data extraction.

Queries PubChem database (110M+ compounds) for chemical properties, similarity, and bioactivity via PUG-REST API or PubChemPy.

Performs DC power flow analysis for transmission networks, including susceptance matrix construction, line flow calculation, and sensitivity analysis.

Runs Python ML workloads on cloud GPUs (A100/T4/A10G) via Modal, handling setup, GPU selection, data management, and results for model training.

Provides best practices and workflows for exoplanet detection and characterization from light curve data, including pipeline planning and method selection.