long-data: Human-AI Vibe Coding Interaction Data (Long Format)

In Nestimate: Network Estimation, Bootstrap, and Higher-Order Analysis

Human-AI Vibe Coding Interaction Data (Long Format)

Description

Coded interaction sequences from 429 human-AI pair programming sessions across 34 projects, in long format.

Usage

human_long

ai_long

Format

Data frames in long format with 9 columns:

message_id

Integer. Turn index.

project

Character. Project identifier (Project_1 .. Project_34).

session_id

Character. Unique session hash.

timestamp

Integer. Unix timestamp for ordering.

session_date

Character. Date of the session (YYYY-MM-DD).

code

Character. Interaction code (17 action labels).

cluster

Character. High-level cluster: Action, Communication, Directive, Evaluative, Metacognitive, or Repair.

code_order

Integer. Order of the code within the session.

order_in_session

Integer. Absolute turn order within the session.

human_long	Human turns only, 10,796 rows
ai_long	AI turns only, 8,551 rows

An object of class data.frame with 10796 rows and 9 columns.

An object of class data.frame with 8551 rows and 9 columns.

Source

Saqr, M. (2026). Human-AI vibe coding interaction study. https://saqr.me/blog/2026/human-ai-interaction-cograph/

Examples

net <- build_network(human_long, method = "tna",
                     action = "code", actor = "session_id",
                     time = "timestamp")
net

Nestimate documentation built on April 20, 2026, 5:06 p.m.