---
name: pymc-probabilistic-programming
description: PyMC for flexible Bayesian modeling
allowed-tools:
  - Bash
  - Read
  - Write
  - Edit
  - Glob
  - Grep
metadata:
  specialization: mathematics
  domain: science
  category: statistical-computing
  phase: 6
---

# PyMC Probabilistic Programming

## Purpose

Provides PyMC capabilities for flexible Bayesian modeling and probabilistic programming in Python.

## Capabilities

- Hierarchical model specification
- Custom distributions
- Gaussian processes
- MCMC and variational inference
- Model diagnostics
- ArviZ integration for visualization

## Usage Guidelines

1. **Model Building**: Use PyMC context managers
2. **Custom Distributions**: Define distributions when needed
3. **Hierarchical Models**: Build proper hierarchical structures
4. **Visualization**: Use ArviZ for diagnostic plots

## Tools/Libraries

- PyMC
- ArviZ
- Theano/PyTensor