Skip to content

Commit 521e05e

Browse files
committed
fix(docs): better formatting
1 parent c9a815a commit 521e05e

File tree

1 file changed

+118
-80
lines changed

1 file changed

+118
-80
lines changed

docs/verifiable-ai-hackathon/0-onboarding-guide.mdx

Lines changed: 118 additions & 80 deletions
Original file line numberDiff line numberDiff line change
@@ -22,31 +22,32 @@ import AddMetaData from "/static/img/verifiable-ai-hackathon/add-metadata.png";
2222

2323
This guide provides all necessary steps to get onboarded for the **Verifiable AI Hackathon**. Participants will set up their **Google Cloud infrastructure**, acquire necessary **API keys**, and deploy their **Confidential Compute VMs**.
2424

25-
## 1. Setup
25+
## 1. Initial Setup
2626

27-
1. **Google Account**
27+
### 1.1. Google Account
2828

29-
- Create or use an existing Google account
30-
- This account will be used for all hackathon resources
29+
- Create or use an existing Google account
30+
- This account will be used for all hackathon resources
3131

32-
2. **Project Registration**
32+
### 1.2. Project Registration
3333

34-
- Submit your email address during registration
35-
- The Flare Infra Team will onboard you to `google-hackathon-project`
34+
- Submit your email address during registration
35+
- The Flare Infra Team will onboard you to `google-hackathon-project`
3636

37-
3. **Gemini API Key**
37+
### 1.3. Gemini API Key
3838

39-
1. Navigate to [Get a Gemini API key in Google AI Studio](https://aistudio.google.com/app/apikey)
40-
2. Click on "Create API key"
41-
3. Select `google-hackathon-project` from the dropdown
42-
4. Click "Create API key in existing project"
39+
1. Navigate to [Get a Gemini API key in Google AI Studio](https://aistudio.google.com/app/apikey)
40+
2. Click on "Create API key"
41+
3. Select `google-hackathon-project` from the dropdown
42+
4. Click "Create API key in existing project"
4343

44-
4. **VM Instance Creation**
45-
- Choose between CLI or Console setup methods below
44+
### 1.4. VM Instance Creation
45+
46+
Choose between CLI or Console setup methods detailed in Section 2.
4647

4748
## 2. VM Instance Setup Methods
4849

49-
### Option A: Using gcloud CLI (Recommended)
50+
### 2.1. Option A: Using gcloud CLI (Recommended)
5051

5152
Follow these steps to set up a GCP instance using Confidential Space with Cloud Logging.
5253

@@ -86,7 +87,7 @@ Follow these steps to set up a GCP instance using Confidential Space with Cloud
8687
- `<ENV_VAR_VALUE1>`: First environment variable value
8788
- Add additional environment variables as needed
8889

89-
### Option B: Using GCP Console
90+
### 2.2. Option B: Using GCP Console
9091

9192
Follow these steps to set up a GCP instance using the web console.
9293

@@ -95,71 +96,108 @@ Follow these steps to set up a GCP instance using the web console.
9596
- Go to the [GCP Console](https://console.cloud.google.com/)
9697
- Click "Create Instance"
9798

98-
{" "}
99-
100-
<img src={CreateInstance} />
101-
102-
2. **Configure Machine**
103-
104-
1. Basic Settings:
105-
106-
- Name: `<project-name>-<team-name>` (e.g., agent-gemini-john)
107-
- Region: us-west2
108-
- Machine Type: n2d-standard-2 (2 vCPU, 1 core, 8 GB memory)
109-
110-
<img src={MachineConfiguration} />
111-
112-
2. Security Settings:
113-
114-
- Service account: confidential-sa
115-
- Access scopes: "Allow full access to Cloud APIs"
116-
- Confidential VM service: Enable
117-
- Shielded VM: Check "Turn on Secure Boot"
118-
119-
<img src={Security} />
120-
121-
3. OS and Storage:
122-
123-
1. Click "Change"
124-
2. Select "Confidential Space Debug 240900"
125-
3. Keep other values default
126-
4. Click "Select"
127-
128-
<img src={OsAndStorage} />
129-
130-
4. Networking:
131-
132-
- Enable "Allow HTTPS traffic"
133-
134-
<img src={Networking} />
135-
136-
5. Advanced Settings:
137-
138-
1. Click "Add Item" under Metadata
139-
2. Add the following items:
140-
141-
```bash
142-
tee-image-reference: <ghcr.io/john/agent-gemini:main>
143-
tee-container-log-redirect: cloud_logging
144-
```
145-
146-
3. Additional API keys should be prefixed with `tee-env-`
147-
148-
<img src={AddMetaData} />
149-
150-
3. **Launch Instance**
99+
<div align="center">
100+
<img
101+
src={CreateInstance}
102+
alt="Create Instance Button"
103+
style={{ maxWidth: "80%", margin: "20px 0" }}
104+
/>
105+
</div>
106+
107+
2. **Configure Machine Settings**
108+
109+
a. **Basic Settings:**
110+
111+
- Name: `<project-name>-<team-name>` (e.g., agent-gemini-john)
112+
- Region: us-west2
113+
- Machine Type: n2d-standard-2 (2 vCPU, 1 core, 8 GB memory)
114+
115+
<div align="center">
116+
<img
117+
src={MachineConfiguration}
118+
alt="Machine Configuration"
119+
style={{ maxWidth: "80%", margin: "20px 0" }}
120+
/>
121+
</div>
122+
123+
b. **Security Settings:**
124+
125+
- Service account: confidential-sa
126+
- Access scopes: "Allow full access to Cloud APIs"
127+
- Confidential VM service: Enable
128+
- Shielded VM: Check "Turn on Secure Boot"
129+
130+
<div align="center">
131+
<img
132+
src={Security}
133+
alt="Security Settings"
134+
style={{ maxWidth: "80%", margin: "20px 0" }}
135+
/>
136+
</div>
137+
138+
c. **OS and Storage:**
139+
140+
1. Click "Change"
141+
2. Select "Confidential Space Debug 240900"
142+
3. Keep other values default
143+
4. Click "Select"
144+
145+
<div align="center">
146+
<img
147+
src={OsAndStorage}
148+
alt="OS and Storage Configuration"
149+
style={{ maxWidth: "80%", margin: "20px 0" }}
150+
/>
151+
</div>
152+
153+
d. **Networking:**
154+
155+
- Enable "Allow HTTPS traffic"
156+
157+
<div align="center">
158+
<img
159+
src={Networking}
160+
alt="Networking Configuration"
161+
style={{ maxWidth: "80%", margin: "20px 0" }}
162+
/>
163+
</div>
164+
165+
e. **Advanced Settings:**
166+
167+
1. Click "Add Item" under Metadata
168+
2. Add the following items:
169+
```bash
170+
tee-image-reference: <ghcr.io/john/agent-gemini:main>
171+
tee-container-log-redirect: cloud_logging
172+
```
173+
3. Additional API keys should be prefixed with `tee-env-`
174+
175+
<div align="center">
176+
<img
177+
src={AddMetaData}
178+
alt="Add Metadata Configuration"
179+
style={{ maxWidth: "80%", margin: "20px 0" }}
180+
/>
181+
</div>
182+
183+
3. **Launch and Monitor Instance**
151184

152185
- Click "Create" to start the VM
153186
- Monitor progress:
187+
1. System logs: Click "Serial Port (console)"
188+
2. Container logs: Click "Logging" after startup
154189

155-
- System logs: Click "Serial Port (console)"
156-
- Container logs: Click "Logging" after startup
157-
158-
<img src={FinalStep} />
190+
<div align="center">
191+
<img
192+
src={FinalStep}
193+
alt="Final Step Configuration"
194+
style={{ maxWidth: "80%", margin: "20px 0" }}
195+
/>
196+
</div>
159197

160-
## Resource Allocation
198+
## 3. Resource Allocation
161199

162-
The hackathon provides the following resource allocations per track:
200+
### Track Allocations
163201

164202
| Track | Prize Share | Compute | Gemini API | BigQuery | Total |
165203
| ------------------ | ----------- | ------- | ---------- | -------- | ----- |
@@ -168,24 +206,24 @@ The hackathon provides the following resource allocations per track:
168206
| AI x DeFi | 35% | $250 | $100 | $400 | $750 |
169207
| Consensus Learning | 20% | $250 | $100 | $300 | $650 |
170208

171-
**Notes:**
209+
**Important Notes:**
172210

173211
- Total credits required: ~$3,000
174212
- Buffer allocation: 100% ($6,000 total)
175213
- Resources monitored via Billing → Budget & Alerts
176214
- Consumption tracked per service
177215

178-
## Service Usage Assumptions
216+
## 4. Service Usage Guidelines
179217

180-
### BigQuery Usage
218+
### 4.1. BigQuery Usage
181219

182220
- 50% team activity at any time
183221
- Free tier (1 TiB) distributed equally
184222
- 48-hour usage period
185223
- Query patterns vary by track
186224
- Storage costs not included
187225

188-
### Gemini API Usage
226+
### 4.2. Gemini API Usage
189227

190228
- Uses Gemini 1.5 Flash pricing
191229
- 50% team activity rate
@@ -194,7 +232,7 @@ The hackathon provides the following resource allocations per track:
194232
- No context caching included
195233
- No Google Search grounding
196234

197-
### Compute Resource Usage
235+
### 4.3. Compute Resource Usage
198236

199237
- Confidential Compute: $0.123/hour per instance
200238
- Total hourly cost: $15.65 (125 teams)

0 commit comments

Comments
 (0)