Skip to content

Commit 281e84d

Browse files
sanu17sanu17
authored
Merge pull request #1280 from BudEcosystem/codex/create-documentation-for-project-module-gx6c0c
fix(docs): enforce trailing-newline and whitespace-safe formatting in Projects docs
2 parents 97aacbc + 2c488e5 commit 281e84d

File tree

8 files changed

+580
-0
lines changed

8 files changed

+580
-0
lines changed

‎master/docs/projects/creating-first-project.mdx‎

Lines changed: 84 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,84 @@
1+
---
2+
title: "Creating Your First Project"
3+
description: "Step-by-step guide to create a project and operationally prepare it"
4+
---
5+
6+
This guide walks through a practical setup flow for your first Bud AI Foundry project.
7+
8+
## Before You Start
9+
10+
Make sure you have permission to create and manage projects in your workspace.
11+
12+
## 1. Create the Project Record
13+
14+
1. Open **Projects**.
15+
2. Click **+ Project**.
16+
3. Fill in:
17+
- **Name**: `Fraud Detection Assistant`
18+
- **Description**: `Project for fraud triage model APIs`
19+
- **Tags**: `risk`, `production`, `llm`
20+
4. Click **Create**.
21+
22+
## 2. Add Team Members
23+
24+
1. Open the new project.
25+
2. Click **Share/Invite**.
26+
3. Add users and assign view/manage scopes.
27+
4. Confirm invited members appear in project access controls.
28+
29+
## 3. Deploy a Model in the Project
30+
31+
1. Go to the **Deployments** tab.
32+
2. Create a deployment and select:
33+
- Model source (for example Hugging Face or cloud provider)
34+
- Target cluster
35+
- Deployment name
36+
3. Save and wait for deployment status to become healthy/ready.
37+
38+
## 4. Configure Deployment Settings
39+
40+
Open deployment **Settings** and configure:
41+
42+
- **Rate limits** to protect capacity
43+
- **Retry/fallback** rules for resiliency
44+
- **Autoscaling** minimum/maximum replicas and trigger metrics
45+
46+
## 5. Add Route Logic
47+
48+
1. Open the **Routes** tab.
49+
2. Create a route and select strategy (for example simple shuffle or least busy).
50+
3. Attach one or more eligible deployments.
51+
4. Save and validate route status.
52+
53+
## 6. Validate Analytics and Observability
54+
55+
1. Open **Analytics** to inspect baseline latency/throughput.
56+
2. Open **Observability** to verify runtime event visibility.
57+
3. Confirm traffic and error patterns are visible before broad rollout.
58+
59+
## 7. Apply Guardrails
60+
61+
1. Open **Guardrails**.
62+
2. Attach required safety or compliance policies.
63+
3. Verify policies are active and aligned with project requirements.
64+
65+
## Completion Checklist
66+
67+
- [x] Project created with metadata
68+
- [x] Team members invited with scoped permissions
69+
- [x] At least one deployment running
70+
- [x] Route configured to active deployment
71+
- [x] Analytics and observability validated
72+
- [x] Guardrails applied
73+
74+
## End-to-End Setup Flow
75+
76+
```mermaid
77+
flowchart TD
78+
A[Create Project] --> B[Add Members]
79+
B --> C[Create Deployment]
80+
C --> D[Configure Settings]
81+
D --> E[Create Route]
82+
E --> F[Validate Analytics & Observability]
83+
F --> G[Enable Guardrails]
84+
```

‎master/docs/projects/guides/deployment-settings-and-autoscaling.mdx‎

Lines changed: 50 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,50 @@
1+
---
2+
title: "Deployment Settings and Autoscaling"
3+
description: "Configure reliability, rate limits, and SLO-aware scaling inside a project"
4+
---
5+
6+
## Overview
7+
8+
Project deployments support operational controls for reliability and cost efficiency.
9+
10+
## Configure Rate Limiting
11+
12+
In deployment **Settings**:
13+
14+
1. Enable **Rate Limit**.
15+
2. Choose an algorithm (fixed, sliding, or token bucket based on availability).
16+
3. Set request thresholds and burst behavior.
17+
4. Save and verify limits in runtime behavior.
18+
19+
## Configure Retry and Fallback
20+
21+
1. Set retry limits for transient failures.
22+
2. Add fallback deployment targets when available.
23+
3. Save and test with controlled failure scenarios.
24+
25+
## Enable Autoscaling
26+
27+
1. Toggle **Autoscaling** on.
28+
2. Set minimum and maximum replicas.
29+
3. Select trigger metrics (for example queue depth or latency-aligned signals).
30+
4. Configure stabilization windows/policies to avoid rapid oscillation.
31+
5. Save and monitor scale events in analytics/observability.
32+
33+
## Best Practices
34+
35+
<Check>Start with conservative max replica values, then scale after observing load.</Check>
36+
<Check>Pair autoscaling thresholds with realistic SLO targets.</Check>
37+
<Check>Use fallback only when secondary deployments are validated.</Check>
38+
<Check>Review settings after model or traffic profile changes.</Check>
39+
40+
## Deployment Controls Flow
41+
42+
```mermaid
43+
flowchart LR
44+
A[Deployment Settings] --> B[Rate Limiting]
45+
A --> C[Retry & Fallback]
46+
A --> D[Autoscaling]
47+
D --> E[Min/Max Replicas]
48+
D --> F[SLO Trigger Metrics]
49+
D --> G[Stabilization Policies]
50+
```

‎master/docs/projects/guides/managing-project-members.mdx‎

Lines changed: 51 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,51 @@
1+
---
2+
title: "Managing Project Members"
3+
description: "Invite collaborators and manage project-level permissions"
4+
---
5+
6+
## Overview
7+
8+
Project member management lets you control who can view and operate project resources.
9+
10+
## Invite Members
11+
12+
1. Open the target project.
13+
2. Click **Share** or **Invite**.
14+
3. Add one or more users.
15+
4. Assign permission scope (view/manage).
16+
5. Save and confirm invite status.
17+
18+
## Update Permissions
19+
20+
1. Open project member management.
21+
2. Locate the user.
22+
3. Change role/scope as needed.
23+
4. Save and verify updated access.
24+
25+
## Remove Members
26+
27+
1. Open project member management.
28+
2. Select a user.
29+
3. Click **Remove**.
30+
4. Confirm the action.
31+
32+
## Recommended Access Model
33+
34+
- Grant **manage** only to platform owners and on-call operators.
35+
- Grant **view** to stakeholders who need visibility but not mutation rights.
36+
- Review membership regularly, especially after team changes.
37+
38+
## Related Docs
39+
40+
- [Project Concepts](/projects/project-concepts)
41+
- [Troubleshooting](/projects/troubleshooting)
42+
43+
## Member Management Workflow
44+
45+
```mermaid
46+
flowchart LR
47+
A[Open Project] --> B[Invite Member]
48+
B --> C[Assign Scope]
49+
C --> D[Review Access]
50+
D --> E[Update or Remove Member]
51+
```

‎master/docs/projects/introduction.mdx‎

Lines changed: 73 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,73 @@
1+
---
2+
title: "Introduction to Projects"
3+
description: "Organize deployments, teams, routes, and governance in a single workspace"
4+
---
5+
6+
## Overview
7+
8+
The **Projects** module is the operational home for delivery teams in Bud AI Foundry. A project groups the assets you build and run—models, deployments, agents, routes, analytics, and guardrails—so teams can work in one governed workspace instead of jumping between tools.
9+
10+
Use projects to define ownership boundaries, manage access, and keep production operations auditable.
11+
12+
## Why Projects Matter
13+
14+
**Unified AI Delivery Workspace**
15+
Track everything for an application in one place: deployment lifecycle, prompt/agent assets, routing, and runtime analytics.
16+
17+
**Collaboration with RBAC**
18+
Project-level permissions control who can view, manage, deploy, edit routes, and configure settings.
19+
20+
**Built-in Governance**
21+
Guardrails, observability, and deletion safeguards help teams enforce policy and reduce production risk.
22+
23+
**Cost and Performance Control**
24+
Project analytics and deployment settings support SLO-aware tuning, autoscaling, and reliability controls.
25+
26+
## Project Lifecycle Flow
27+
28+
```mermaid
29+
flowchart LR
30+
A[Create Project] --> B[Invite Members]
31+
B --> C[Deploy Models]
32+
C --> D[Configure Routes]
33+
D --> E[Monitor Analytics & Observability]
34+
E --> F[Apply Guardrails]
35+
F --> G[Iterate with SLO-aware Scaling]
36+
```
37+
38+
## Core Areas Inside a Project
39+
40+
| Area | What you can do |
41+
|------|------------------|
42+
| **Deployments** | Create endpoints, inspect status, tune scaling and limits |
43+
| **Agents** | Manage project-linked agents and prompts |
44+
| **Routes** | Define traffic strategy across one or more deployments |
45+
| **Analytics** | Track latency, throughput, and utilization trends |
46+
| **Clusters** | View target infrastructure and placement context |
47+
| **Observability** | Inspect inference activity and operational signals |
48+
| **Guardrails** | Apply and monitor safety/compliance controls |
49+
50+
## Typical Project Workflow
51+
52+
1. **Create a project** with clear name, description, and tags.
53+
2. **Invite collaborators** with view/manage permissions.
54+
3. **Deploy models** into project-scoped endpoints.
55+
4. **Configure routing** and reliability controls.
56+
5. **Monitor analytics** and tune scaling/performance.
57+
6. **Harden governance** with guardrails and access reviews.
58+
59+
## Getting Started
60+
61+
<CardGroup cols={3}>
62+
<Card title="Quick Start" icon="play" href="/projects/quickstart">
63+
Create and configure your first project in minutes
64+
</Card>
65+
66+
<Card title="Project Concepts" icon="book" href="/projects/project-concepts">
67+
Understand structure, permissions, and module tabs
68+
</Card>
69+
70+
<Card title="Step-by-Step Tutorial" icon="graduation-cap" href="/projects/creating-first-project">
71+
Full walkthrough from project setup to deployment readiness
72+
</Card>
73+
</CardGroup>

‎master/docs/projects/project-concepts.mdx‎

Lines changed: 74 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,74 @@
1+
---
2+
title: "Project Concepts"
3+
description: "Understand project structure, permissions, and operational boundaries"
4+
---
5+
6+
## Project as an Operational Boundary
7+
8+
A project is the primary boundary for organizing AI assets and operations in Bud AI Foundry. Most day-to-day activities—deployment lifecycle, routing, analytics, and safety controls—are scoped to a project.
9+
10+
## Core Concepts
11+
12+
### 1. Project Metadata
13+
14+
Each project has a name, icon, description, and tags. Metadata helps teams discover and classify workloads across environments.
15+
16+
### 2. Permission Scope
17+
18+
Project permissions determine who can:
19+
20+
- View project assets
21+
- Manage project settings and members
22+
- Create/edit/delete deployments and routes (depending on role scope)
23+
24+
### 3. Tab-Based Functional Areas
25+
26+
Project detail pages are organized into focused tabs for operational clarity:
27+
28+
- **Deployments**: endpoint lifecycle and runtime settings
29+
- **Agents**: prompt- and agent-centric assets
30+
- **Routes**: traffic distribution strategy
31+
- **Analytics**: performance and capacity trends
32+
- **Clusters**: infrastructure placement context
33+
- **Observability**: runtime traces, logs, and request signals
34+
- **Guardrails**: policy and safety controls
35+
36+
### 4. Safe Lifecycle Management
37+
38+
Project deletion is intentionally guarded. If active deployments are present, deletion can be blocked until dependencies are removed or paused.
39+
40+
### 5. SLO-Aware Scaling
41+
42+
Deployment settings can be tuned with SLO-aligned autoscaling signals (for example latency or queue-related metrics), allowing teams to balance performance and cost.
43+
44+
## How Projects Relate to Other Modules
45+
46+
| Module | Relationship to Projects |
47+
|--------|---------------------------|
48+
| **Models** | Models are selected and deployed into project-scoped endpoints |
49+
| **Deployments** | Deployments are managed and governed within projects |
50+
| **Routes** | Routing policies are configured from project context |
51+
| **Observability** | Monitoring is filtered by project for faster diagnostics |
52+
| **User Management** | Access is granted to project resources via RBAC |
53+
54+
## Best Practices
55+
56+
<Check>Use one project per product, tenant, or major workload boundary.</Check>
57+
<Check>Keep project tags consistent for reporting and governance.</Check>
58+
<Check>Grant manage permissions to a minimal set of operators.</Check>
59+
<Check>Review analytics and observability before scaling or route changes.</Check>
60+
<Check>Enable guardrails early in lifecycle, not only after incidents.</Check>
61+
62+
## Concept Map
63+
64+
```mermaid
65+
flowchart LR
66+
A[Project Boundary] --> B[Permissions]
67+
A --> C[Operational Tabs]
68+
A --> D[Governance]
69+
C --> C1[Deployments]
70+
C --> C2[Routes]
71+
C --> C3[Analytics]
72+
D --> D1[Guardrails]
73+
D --> D2[Deletion Safeguards]
74+
```

0 commit comments

Comments
 (0)