Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
135 changes: 135 additions & 0 deletions docs/user-guide/analytics/ai-champions-leaderboard.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,135 @@
---
id: ai-champions-leaderboard
title: AI Champions Leaderboard
sidebar_label: AI Champions Leaderboard
sidebar_position: 1
pagination_prev: user-guide/analytics/analytics-overview
pagination_next: null
---

# AI Champions Leaderboard

The AI Champions Leaderboard ranks platform users by their overall AI engagement score,
calculated across six dimensions of platform activity. It gives platform admins a structured
view of AI adoption patterns and helps identify top contributors across the organization.

:::info
The full leaderboard is visible to [**platform admins**](../getting-started/glossary.md#platform-admin) only. Regular users can view their
own profile and score, but cannot see other users' rankings.
:::

## Accessing the Leaderboard

1. Click the **Analytics** icon in the left sidebar navigation.
2. Select the **Leaderboard** tab at the top of the Analytics Dashboard.

Admins see the full ranked table with filters and aggregated statistics. Regular users land
directly on their personal score card.

![AI Champions Leaderboard overview](./images/leaderboard-overview.png)

## Scoring Model

Each user receives a total score from **0 to 100**, calculated as a weighted sum of six
dimension scores.

| # | Dimension | Weight | What it measures |
| --- | ------------------------- | ------ | ------------------------------------------------------------- |
| D1 | Core Platform Usage | 20% | Active days, conversations, assistants used, platform breadth |
| D2 | Core Platform Creation | 20% | Assistants, skills, and data sources created; asset maturity |
| D3 | Workflow Usage | 10% | Workflow execution volume, success rate, repeat usage |
| D4 | Workflow Creation | 10% | Workflows built, structural complexity, originality |
| D5 | CLI & Agentic Engineering | 30% | CLI sessions, repositories, code output, delivery efficiency |
| D6 | Impact & Knowledge | 10% | Asset reuse by others, knowledge sharing, kata completions |

D5 carries the highest weight, reflecting that agentic, code-level AI usage is the most
advanced form of platform engagement.

:::note
Users with a total score of **0** are excluded from the leaderboard entirely. Only active
users appear in rankings.
:::

## Tiers

Users are assigned a tier based on their total score:

| Tier | Min Score | Description |
| ---------------- | --------- | --------------------------------------------------- |
| **Pioneer** | 80+ | Exceptional breadth and depth across all dimensions |
| **Expert** | 65–79 | Strong engagement across most platform areas |
| **Advanced** | 45–64 | Consistent usage with meaningful creation activity |
| **Practitioner** | 25–44 | Regular platform user with growing engagement |
| **Newcomer** | < 25 | Early-stage user just getting started |

## Intent Classification

In addition to a tier, each user is assigned an **intent** label that reflects their primary
usage pattern:

| Intent | Description |
| ---------------------- | -------------------------------------------------------------------------- |
| **SDLC Unicorn** | Strong across platform usage, creation, workflows, and CLI engineering |
| **Developer** | CLI-dominant usage — primarily uses CodeMie through the command-line agent |
| **Workflow Architect** | Primarily builds and creates workflows |
| **Workflow User** | Primarily executes and consumes workflows |
| **Platform Builder** | Focused on creating assistants, skills, and data sources |
| **AI User** | Primarily uses the platform through web conversations and assistants |
| **Explorer** | Very low overall activity; early exploration stage |

## Leaderboard Views

The leaderboard supports three time-based views, selectable from the view switcher at the
top of the tab:

| View | Description |
| ----------------------- | --------------------------------------------- |
| **Current 30 days** | Rolling 30-day window, recomputed nightly |
| **Monthly champions** | Final snapshot for a completed calendar month |
| **Quarterly champions** | Final snapshot for a completed quarter |

Monthly and quarterly views are archived snapshots — they reflect the state at the end of
that period and do not change.

## Admin Capabilities

Platform admins have access to additional leaderboard functionality:

### Browsing and Filtering

The leaderboard table supports:

- **Search** by user name or email
- **Filter** by tier or intent
- **Sort** by rank, total score, username, or tier level

### User Detail View

Click any user row to open their detailed profile, which shows:

- Scores and component breakdowns for all six dimensions
- Assigned tier and intent
- Active projects

### Leaderboard Summary

The top of the Leaderboard tab displays aggregated statistics:

- Total scored users
- Average score across the organization
- User count per tier
- Top score in the current period

## Configuration

The leaderboard feature is controlled by the following environment variables, set by the
platform administrator:

| Variable | Default | Description |
| ------------------------- | ----------- | ------------------------------------------- |
| `LEADERBOARD_ENABLED` | `false` | Enables the nightly computation job |
| `LEADERBOARD_SCHEDULE` | `0 2 * * *` | Cron expression for the scheduled run (UTC) |
| `LEADERBOARD_PERIOD_DAYS` | `30` | Rolling window size in days |

If `LEADERBOARD_ENABLED` is `false`, the leaderboard tab is not available and the nightly
job does not run. Contact your platform administrator to enable this feature.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
12 changes: 12 additions & 0 deletions docs/user-guide/analytics/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -131,3 +131,15 @@ The Analytics Dashboard features a powerful filtering system that allows you to
- **Combined Filtering**: All active filters work together to refine data
- **Persistent Selections**: Filter choices remain active as you navigate between dashboard sections
- **Clear All**: Reset all filters with a single action using the "Clear all" button

## AI Champions Leaderboard

The Analytics Dashboard includes an **AI Champions Leaderboard** tab that ranks users by
their overall AI platform engagement score. Scores are calculated across six dimensions —
from daily platform usage to CLI-based agentic engineering — and users are grouped into
tiers from Newcomer to Pioneer.

Platform admins can browse the full ranked leaderboard with filters and drill into individual
user profiles. Regular users can view their own score and tier.

See [AI Champions Leaderboard](./ai-champions-leaderboard.md) for the full overview.
4 changes: 4 additions & 0 deletions docs/user-guide/getting-started/glossary.md
Original file line number Diff line number Diff line change
Expand Up @@ -194,6 +194,10 @@ A workspace within the CodeMie platform that delineates and isolates working env

**Important**: Project names must be in lowercase only. Uppercase characters are not allowed in project names.

### Platform Admin

A user with platform-wide administrative privileges in CodeMie, identified by the `isAdmin = true` flag. Platform Admins can create projects, manage all users and integrations, access platform-wide analytics data, and perform all administrative operations across the entire platform, regardless of project membership.

### Project Admin

A user with administrative privileges within the CodeMie platform. Project Admins have extended functionality including the ability configure Project Integrations (available to all team members), view all project assistants and workflows and manage project-level settings. Only Admin roles or user with attribute applications_admin have access to these administrative functions.
Expand Down
25 changes: 25 additions & 0 deletions faq/how-is-my-ai-champions-leaderboard-score-calculated.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,25 @@
# How is my AI Champions Leaderboard score calculated?

Your score is a weighted sum of six dimension scores, each normalized to a 0–1 range and
multiplied by the dimension's weight. The final result is scaled to 0–100.

| Dimension | Weight |
| ----------------------------- | ------ |
| D1: Core Platform Usage | 20% |
| D2: Core Platform Creation | 20% |
| D3: Workflow Usage | 10% |
| D4: Workflow Creation | 10% |
| D5: CLI & Agentic Engineering | 30% |
| D6: Impact & Knowledge | 10% |

CLI and agentic engineering (D5) carries the highest weight at 30%, reflecting that
command-line and agent-based AI usage represents the most advanced form of platform
engagement.

Each dimension is itself composed of weighted sub-components — for example, D1 counts
active days, number of conversations, assistants used, and platform feature breadth. To
improve your score, focus on expanding usage across more dimensions, especially D5.

## Sources

- [AI Champions Leaderboard](https://codemie-ai.github.io/docs/user-guide/analytics/ai-champions-leaderboard)
17 changes: 17 additions & 0 deletions faq/what-is-the-ai-champions-leaderboard.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,17 @@
# What is the AI Champions Leaderboard?

The AI Champions Leaderboard is a feature in the Analytics Dashboard that ranks all platform
users by their overall AI engagement score. Scores range from 0 to 100 and are calculated
across six dimensions of platform activity: core usage, asset creation, workflow usage,
workflow creation, CLI and agentic engineering, and impact and knowledge sharing.

Users are grouped into five tiers — Newcomer, Practitioner, Advanced, Expert, and Pioneer —
based on their total score. Each user also receives an intent label (such as Developer,
Platform Builder, or SDLC Unicorn) that reflects their dominant usage pattern.

The full leaderboard is visible to platform admins only. Regular users can view their own
score, tier, and dimension breakdown in their personal profile.

## Sources

- [AI Champions Leaderboard](https://codemie-ai.github.io/docs/user-guide/analytics/ai-champions-leaderboard)
9 changes: 7 additions & 2 deletions sidebars.ts
Original file line number Diff line number Diff line change
Expand Up @@ -301,9 +301,14 @@ const sidebars: SidebarsConfig = {
],
},
{
type: 'doc',
id: 'user-guide/analytics/analytics-overview',
type: 'category',
label: 'Analytics',
link: {
type: 'doc',
id: 'user-guide/analytics/analytics-overview',
},
collapsed: true,
items: ['user-guide/analytics/ai-champions-leaderboard'],
},
{
type: 'category',
Expand Down
Loading