Implementation:DataExpert io Data engineer handbook Statsig Get experiment

Overview

Statsig Get Experiment is a Wrapper Doc implementation that documents the use of the external Statsig Python SDK to assign users to experiment variants and retrieve variant-specific parameters. This is the concrete realization of the Principle:DataExpert_io_Data_engineer_handbook_Experiment_User_Assignment principle.

Type

Wrapper Doc (external Statsig SDK)

Source

server.py:L50-76

Signature

statsig.get_experiment(user: StatsigUser, experiment_name: str) -> DynamicConfig

Then, to retrieve individual parameters from the returned config:

DynamicConfig.get(param_name: str, default: str) -> str

User Creation

Users are identified by creating a StatsigUser object. The application supports two identity strategies:

# Default: hash the client IP address for a pseudo-anonymous but deterministic ID
StatsigUser(str(hash(request.remote_addr)))

# Alternative: random ID via ?random query parameter for full anonymity
StatsigUser(random_id)

IP hashing (default): Uses str(hash(request.remote_addr)) to create a deterministic user ID from the client's IP address. The same IP always produces the same hash, ensuring consistent variant assignment across visits.
Random ID (via ?random query parameter): Generates a random identifier, useful for testing or when IP-based bucketing is not desired.

Experiment Configuration

Experiment name: "button_color_v3"
Retrieved parameters:

Parameter Name	Default Value	Description
`"Button Color"`	`"blue"`	The color of the call-to-action button displayed to the user
`"Paragraph Text"`	`"Data Engineering Boot Camp"`	The headline or paragraph text shown on the page

Imports

from statsig import statsig
from statsig.statsig_user import StatsigUser

Inputs and Outputs

Direction	Name	Type	Description
Input	user	StatsigUser	A user object constructed from either a hashed IP address or a random ID
Input	experiment_name	str	The name of the experiment to evaluate (e.g., `"button_color_v3"`)
Output	DynamicConfig	DynamicConfig	A configuration object containing the variant parameters assigned to this user. Individual values are retrieved via `.get(param_name, default)`.

Behavior

The SDK evaluates the user against the experiment's bucketing rules to determine which variant the user belongs to
The returned DynamicConfig contains all parameter values for the assigned variant
If the experiment does not exist or the user is not eligible, the default values are returned
Assignment is deterministic — the same StatsigUser ID always receives the same variant

External Reference

Statsig Python SDK Documentation

Related Pages

Metadata

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment