feat(docs): update to include pezzo proxy

pezzolabs · Nov 24, 2023 · 5f16125 · 5f16125
1 parent f55e6fb
commit 5f16125
Show file tree

Hide file tree

Showing 10 changed files with 166 additions and 159 deletions.
diff --git a/docs/client/request-caching.mdx b/docs/client/request-caching.mdx
@@ -11,8 +11,10 @@ Utilizing caching can sometimes reduce your development costs and execution time
 
 ## Usage
 
-To enable caching, simply set `cache: enabled` in the Pezzo Options parameter. Here is an example:
+To enable caching, simply set the `X-Pezzo-Cache-Enabled: true` header. Here is an example:
 
+<Tabs>
+  <Tab title="Node.js">
 ```ts
 const response = await openai.chat.completions.create({
   model: "gpt-3.5-turbo",
@@ -23,9 +25,30 @@ const response = await openai.chat.completions.create({
     }
   ]
 }, {
-  cache: true
+  headers: {
+    "X-Pezzo-Cache-Enabled": true,
+  }
 });
+
+```
+  </Tab>
+  <Tab title="Python">
+```py
+chat_completion = openai.chat.completions.create(
+  model="gpt-3.5-turbo",
+  messages=[
+    {
+      "role": "user",
+      "content": "Tell me 5 fun facts about yourself",
+    }
+  ],
+  headers={
+    "X-Pezzo-Cache-Enabled": "true"
+  }
+)
 ```
+  </Tab>
+</Tabs>
 
 ## Cached Requests in the Console
 

diff --git a/docs/introduction/docker-compose.mdx → docs/deployment/docker-compose.mdx b/docs/introduction/docker-compose.mdx → docs/deployment/docker-compose.mdx
@@ -1,5 +1,5 @@
 ---
-title: "Running With Docker Compose"
+title: "Docker Compose"
 description: "Learn how to run the full Pezzo stack locally with Docker Compose."
 ---
 

diff --git a/docs/introduction/tutorial-observability/overview.mdx b/docs/introduction/tutorial-observability/overview.mdx
@@ -3,30 +3,6 @@ title: "Tutorial: Observability"
 description: "In just a few lines of code, monitor your AI operations seamlessly."
 ---
 
-<Tip>
-This tutorial is for users who want to use Pezzo for observability and monitoring only.
-
-If you also want to use Pezzo to manage your prompts, version control and prompt deployment, check out the [Prompt Management tutorial](/introduction/tutorial-prompt-management/overview).
-</Tip>
-
-**Prefer a video tutorial?** We've prepared a 5-minute video for you! If you want to see the code example, [it's available on Codesandbox](https://codesandbox.io/p/sandbox/pezzo-example-observability-6d2qp6?file=%2Fsrc%2Fapp.ts%3A1%2C1).
-
-
-<div style={{
-  position: "relative",
-  paddingBottom: "62.5%",
-  height: 0
-}}>
-  <iframe
-    src="https://www.loom.com/embed/dd2276e912834b6c8d67f1355e449cae?sid=d6cea226-90bc-4586-ab6c-7150369b06c9"
-    frameborder="0"
-    webkitallowfullscreen
-    mozallowfullscreen
-    allowfullscreen
-    style={{ position: "absolute", top: 0, left: 0, width: "100%", height: "100%" }}
-  ></iframe>
-</div>
-
 ## What you'll learn
 
 You're going to learn how to easily use Pezzo to supercharge your AI operations with monitoring and observability. It takes just a few lines of code!
@@ -37,31 +13,25 @@ You're going to learn how to easily use Pezzo to supercharge your AI operations
   Cloud](https://app.pezzo.ai).
 </Note>
 
-## Install depdendencies
-
-Install the Pezzo Client and the OpenAI SDK:
-
-```bash
-npm i @pezzo/client openai
-```
-
-## Making calls to OpenAI
+## Using Pezzo with OpenAI
 
 Here is a code example:
 
-```ts app.ts
-import { Pezzo, PezzoOpenAI } from "@pezzo/client";
+<Tabs>
+  <Tab title="Node.js">
+```ts
+import OpenAI from "openai";
 
 // Initialize the Pezzo client
-const pezzo = new Pezzo({
-  apiKey: "<Your Pezzo API key>",
-  projectId: "<Your Pezzo project ID>",
-  environment: "Production",
+const openai = new OpenAI({
+  baseURL: "https://proxy.pezzo.ai/openai/v1",
+  defaultHeaders: {
+    "X-Pezzo-Api-Key": "<Your API Key>",
+    "X-Pezzo-Project-Id": "<Your Project ID>",
+    "X-Pezzo-Environment": "Production",
+  }
 });
 
-// Initialize the OpenAI client
-const openai = new PezzoOpenAI(pezzo);
-
 async function main() {
   // Make calls to the OpenAI API as you normally would!
   const completion = await openai.chat.completions.create(
@@ -71,37 +41,48 @@ async function main() {
       messages: [
         {
           role: "user",
-          content: "Tell me {numFacts} fun facts about {topic}",
+          content: "Tell me 5 fun facts about yourself",
         },
       ],
-    },
-    {
-      variables: {
-        // You can define variables that will be interpolated during execution.
-        numFacts: 3,
-        topic: "Artificial Intelligence",
-      },
-      properties: {
-        // You can optionally specify custom properties that will be associated with the request.
-        someProperty: "someValue",
-      },
     }
   );
 }
 
 main();
-
 ```
+  </Tab>
+  <Tab title="Python">
+```py
+import openai
+
+openai.base_url = "https://proxy.pezzo.ai/openai/v1"
+openai.default_headers = {
+  "X-Pezzo-Api-Key": "<Your API Key>",
+  "X-Pezzo-Project-Id": "<Your Project ID>",
+  "X-Pezzo-Environment": "Production"
+}
 
-[Click here to run this example on CodeSandbox](https://codesandbox.io/p/sandbox/pezzo-example-observability-6d2qp6?file=%2Fsrc%2Fapp.ts%3A1%2C1)
+chat_completion = openai.chat.completions.create(
+  model="gpt-3.5-turbo",
+  messages=[
+    {
+        "role": "user",
+        "content": "Tell me 5 fun facts about yourself",
+    }
+  ]
+)
+```
+  </Tab>
+</Tabs>
 
 **Let's explain what's going on here:**
 
-- First, we initialize the Pezzo client and the OpenAI client. We pass the Pezzo client to the OpenAI client so it can use it to fetch the prompt.
-- Then, we make a call to the OpenAI API as we normally would.
-- (Optional) We specify additional parameters in the second argument, these are `variables` and `properties`.
+We import and instantiate the OpenAI client with a few additional parameters. First, the `baseURL` which tells the OpenAI client to proxy the request through Pezzo.
 
-The result will go directly to OpenAI and the response will be reported to Pezzo.
+Then, we set a few default headers that will be present at any request made to OpenAI. These are:
+- `X-Pezzo-Api-Key` - Your Pezzo API key. You can find it in your Organization page in [Pezzo Cloud](https://app.pezzo.ai).
+- `X-Pezzo-Project-Id` - The ID of the project you want to use. You can find it anywhere [Pezzo Cloud](https://app.pezzo.ai).
+- `X-Pezzo-Environment` - The name of the environment to use. By default, any Pezzo project is automatically created with a `Production` environment.
 
 ## Monitoring Requests
 
@@ -112,3 +93,15 @@ If you want to learn more about Pezzo's observability feature, check out the [Ob
 <Frame caption="Pezzo Project Requests View">
   <img src="/platform/observability/requests-view.png" />
 </Frame>
+
+# Next Steps
+
+<CardGroup cols={2}>
+  <Card
+    title="Request Caching"
+    icon="bolt"
+    href="/client/request-caching"
+  >
+    Save up to 90% of your AI costs with Pezzo's request caching.
+  </Card>
+</CardGroup>
diff --git a/docs/introduction/tutorial-prompt-management/overview.mdx b/docs/introduction/tutorial-prompt-management/overview.mdx
@@ -8,23 +8,6 @@ This tutorial is for users who want to manage their AI operations in Pezzo end-t
 If you wish to only use Pezzo for monitoring and observability, check out the [Observability Tutorial](/introduction/tutorial-observability/overview).
 </Tip>
 
-**Prefer a video tutorial?** We've prepared a 5-minute video for you! If you want to see the code example, [it's available on Codesandbox](https://codesandbox.io/p/sandbox/pezzo-example-prompt-management-qv5f86?file=%2Fsrc%2Fapp.ts%3A1%2C1).
-
-<div style={{
-  position: "relative",
-  paddingBottom: "62.5%",
-  height: 0
-}}>
-  <iframe
-    src="https://www.loom.com/embed/4d7fe830a03943c190a22fbe4704926f?sid=d0312d90-9b34-46d2-bdbd-938911e2ac76"
-    frameborder="0"
-    webkitallowfullscreen
-    mozallowfullscreen
-    allowfullscreen
-    style={{ position: "absolute", top: 0, left: 0, width: "100%", height: "100%" }}
-  ></iframe>
-</div>
-
 ## What you'll learn
 You're going to learn how to manage your AI prompts with Pezzo, so you can streamline delivery and collaborate with your team. This includes:
 
@@ -149,7 +132,7 @@ npm i @pezzo/client openai
 
 Here is a code example:
 
-```ts app.ts
+```ts Node.js
 import { Pezzo, PezzoOpenAI } from "@pezzo/client";
 
 // Initialize the Pezzo client

diff --git a/docs/introduction/what-is-pezzo.mdx b/docs/introduction/what-is-pezzo.mdx
@@ -22,24 +22,17 @@ Pezzo is a powerful open-source toolkit designed to streamline the process of AI
 # Next Steps
 <CardGroup cols={2}>
   <Card
-    title="Observability"
+    title="Tutorial: Observability & Monitoring"
     icon="eye"
-    href="/platform/observability/overview"
+    href="/introduction/tutorial-observability/overview"
   >
-    Learn about Pezzo's robust observability features.
+    Learn about Pezzo's robust observability & monitoring features.
   </Card>
   <Card
-    title="Prompt Management"
+    title="Tutorial: Prompt Management"
     icon="wrench"
-    href="/platform/prompt-management/overview"
+    href="/introduction/tutorial-prompt-management/overview"
   >
     Learn how you can streamline your AI delivery with Pezzo.
   </Card>
-  <Card
-    title="Recipe: OpenAI With Pezzo"
-    icon="code"
-    href="/client/integrations/openai"
-  >
-    Get started with Pezzo and OpenAI in 5 minutes.
-  </Card>
 </CardGroup>
diff --git a/docs/mint.json b/docs/mint.json
@@ -36,35 +36,25 @@
       "group": "Getting Started",
       "pages": [
         "introduction/what-is-pezzo",
-        "introduction/tutorial-prompt-management/overview",
         "introduction/tutorial-observability/overview",
-        "introduction/docker-compose"
+        "introduction/tutorial-prompt-management/overview"
       ]
     },
     {
-      "group": "Observability",
+      "group": "Features",
       "pages": [
+        "platform/proxy/overview",
         "platform/observability/overview",
-        "platform/observability/requests",
-        "platform/observability/metrics"
-      ]
-    },
-    {
-      "group": "Prompt Management",
-      "pages": [
-        "platform/prompt-management/overview",
+        "client/request-caching",
         "platform/prompt-management/environments",
         "platform/prompt-management/prompt-editor",
         "platform/prompt-management/versioning-and-deployments"
       ]
     },
     {
-      "group": "Pezzo SDK",
+      "group": "Deployment",
       "pages": [
-        "client/pezzo-client-node",
-        "client/pezzo-client-python",
-        "client/integrations/openai",
-        "client/request-caching"
+        "deployment/docker-compose"
       ]
     },
     {

diff --git a/docs/platform/observability/metrics.mdx b/docs/platform/observability/metrics.mdx
diff --git a/docs/platform/observability/overview.mdx b/docs/platform/observability/overview.mdx
@@ -1,10 +1,47 @@
 ---
-title: 'Observability Overview'
-sidebarTitle: 'Overview'
+title: 'Monitoring & Observability'
 ---
 Pezzo enabels you to easily observe your Geneartive AI operations. This includes:
 
 - **Traces and Requests** - Pezzo automatically traces your operations and requests and provides you with a detailed view of the execution of your operations.
 - **Advanced Filtering** - Filter your traces and requests by any field, including custom fields (e.g. user ID, correlation ID, etc).
 - **Metrics** - Pezzo automatically collects metrics from your LLM calls (execution time, cost, error rate, and more) and provides you with detailed dashboards.
-- **Alerts** *(Coming Soon)* - Define alerts on your metrics and get notified when something goes wrong.
+- **Alerts** *(Coming Soon)* - Define alerts on your metrics and get notified when something goes wrong.
+
+Observe your Generative AI operations with Pezzo.
+
+# Requests View
+
+Every project on Pezzo has a dedicated Requests view. This view shows you all the requests that have been made to your project.
+
+<Frame caption="Pezzo Project Requests View">
+  <img src="/platform/observability/requests-view.png" />
+</Frame>
+
+## Filters
+
+You can filter requests by various criteria. For example, you can filter by the status of the request, by date/time, or by the cost of execution.
+
+<Frame caption="Filters">
+  <img src="/platform/observability/filters.png" />
+</Frame>
+
+## Inspector
+
+You can inspect the details of any request by clicking on it. This will open the Inspector view.
+
+<Frame caption="Request Inspector">
+  <img src="/platform/observability/request-inspector.png" />
+</Frame>
+
+# Prompt Metrics
+
+<Tip>
+If you feel that there are metrics that are missing, please let us know by [creating an issue on GitHub](https://github.com/pezzolabs/pezzo/issues).
+</Tip>
+
+If you manage your prompts with Pezzo, you are able to view insights and metrics for a specific prompt.
+
+<Frame caption="Prompt Metrics">
+  <img src="/platform/observability/prompt-metrics.png" />
+</Frame>