Deployed bfc44b8 with MkDocs version: 1.6.1

prerequisites/index.html

@@ -596,6 +596,7 @@ <h1>4. Demo Prerequisites</h1>
 <ul>
 <li>A Dynatrace SaaS account (<a href="https://dynatrace.com/trial" target="_blank">free trial</a>)</li>
 <li>An OpenAI account with credit added</li>
+<li>For the RAG version, a (free) PineCone account</li>
 </ul>
 <h2 id="why-is-a-paid-openai-account-required">Why is a Paid OpenAI Account Required?<a class="headerlink" href="#why-is-a-paid-openai-account-required" title="Permanent link">#</a></h2>
 <p>OpenAI / ChatGPT severely limits the ability for API access if you do not have credit. Adding a small amount of credit ($2-$3) is the best way to make this (and all other ChatGPT demos) run smoothly.</p>

resources/index.html

@@ -524,7 +524,8 @@
 
 <h1 id="resources">Resources<a class="headerlink" href="#resources" title="Permanent link">#</a></h1>
 <ul>
-<li><a href="https://github.com/dynatrace-perfclinics/obslab-release-validation" target="_blank">This repository and documentation on GitHub</a></li>
+<li><a href="https://dynatr.ac/3XKxKEC">Official Documentation</a></li>
+<li><a href="https://github.com/Dynatrace/obslab-llm-observability" target="_blank">This repository and documentation on GitHub</a></li>
 <li><a href="https://info.dynatrace.com/apac-all-wb-ensure-ai-project-success-with-ai-observability-24973-registration.html" target="_blank">LLM Observability On-Demand Webinar (Video)</a></li>
 </ul>
 

standard-rag-differences/index.html

@@ -574,15 +574,22 @@
 
   <h1>3. Standard vs. RAG Version</h1>
 
-<p>This demo is available in two flavours.</p>
+<p>This demo is available in two flavors: <em>standard</em> and <em>RAG</em>.</p>
 <p>The "standard" demo uses OpenAI's ChatGPT (coupled with an on-cluster Weaviate cache) to look up destination advice for any destination.</p>
-<p>The "RAG" version (available <a href="https://github.com/dynatrace-perfclinics/obslab-llm-observability/tree/rag" target="_blank">on the rag branch</a>) will <strong>only</strong> produce destination advice for places the system has explicitly been trained on (the files in the <a href="https://github.com/dynatrace-perfclinics/obslab-llm-observability/tree/rag/destinations" target="_blank">destinations folder on the <code>rag</code> branch</a>). Namely, <code>Bali</code> and <code>Sydney</code>.</p>
-<p>This is achieved by:</p>
+<p>The "Retrieval-Augmented Generation" (RAG) version (available <a href="https://github.com/dynatrace-perfclinics/obslab-llm-observability/tree/ollama-pinecone" target="_blank">on the ollama-pinecone branch</a>) mimicks training an LLM on an internal knowledgebase. 
+It will produce custom destination advice for places the system has explicitly been trained on (the files in the <a href="https://github.com/dynatrace-perfclinics/obslab-llm-observability/tree/ollama-pinecone/destinations" target="_blank">destinations folder</a>).
+Namely, <code>Bali</code> and <code>Sydney</code>. For other locations, the model will provide an answer based on its own knowledge.
+It is based on <a href="https://dt-url.net/l843uge">Ollama</a> and uses <a href="https://dt-url.net/0323urx">PineCone</a> as a Vector database. The RAG pipeline is built using <a href="https://dt-url.net/6d03u7j">LangChain</a>. </p>
+<p>The RAG version of the demo mimicks training an LLM on an internal knowledgebase.</p>
+<p><img alt="" src="https://dt-cdn.net/images/architecture-rag-1269-e11f226cf6.jpg" /></p>
+<p>When the application starts, files inside the <a href="https://github.com/dynatrace-perfclinics/obslab-llm-observability/tree/ollama-pinecone/destinations" target="_blank">destinations</a> folder are read, processed, and stored in PineCone for later lookup.
+Afterwards, each request goes through the LangChain RAG pipeline, which performs the following steps:</p>
 <ul>
-<li><a href="https://github.com/dynatrace-perfclinics/obslab-llm-observability/blob/a893c0e8e93b29a0ca1b5482cb0589f9bce0b4cc/app.py#L79" target="_blank">Reading each file from disk when the app starts</a></li>
-<li>Sending the contents of the <a href="https://github.com/dynatrace-perfclinics/obslab-llm-observability/tree/rag/destinations" target="_blank">bali and sydney HTML pages</a> along with each request and <a href="https://github.com/dynatrace-perfclinics/obslab-llm-observability/blob/a893c0e8e93b29a0ca1b5482cb0589f9bce0b4cc/app.py#L100" target="_blank">explicitly telling the model to only use the information provided in those documents</a>.</li>
+<li>It contacts Ollama to produce an embedding of the user input</li>
+<li>With the embedding, reach out to PineCone to find documents relevant to the user input</li>
+<li>Use the documents to perform prompt engineering and send it to Ollama to produce the travel recommendation </li>
+<li>Prosess the answer received </li>
 </ul>
-<p>The RAG version of the demo mimicks training an LLM on an internal knowledgebase.</p>
 <h2 id="click-here-to-continue-with-the-exercise"><a href="../prerequisites/">&gt;&gt; Click here to continue with the exercise</a><a class="headerlink" href="#click-here-to-continue-with-the-exercise" title="Permanent link">#</a></h2>
 
 

startup/index.html

@@ -645,7 +645,7 @@ <h2 id="time-to-fire-it-up">🆙 Time to Fire it up<a class="headerlink" href="#
 <h3 id="launch-standard-version">Launch Standard Version<a class="headerlink" href="#launch-standard-version" title="Permanent link">#</a></h3>
 <p><a href="https://codespaces.new/dynatrace-perfclinics/obslab-llm-observability?ref=main" target="_blank"><img alt="Open &quot;standard&quot; version in GitHub Codespaces" src="https://github.com/codespaces/badge.svg" /></a></p>
 <h3 id="launch-rag-version">Launch RAG Version<a class="headerlink" href="#launch-rag-version" title="Permanent link">#</a></h3>
-<p><a href="https://codespaces.new/dynatrace-perfclinics/obslab-llm-observability?ref=rag" target="_blank"><img alt="Open &quot;RAG&quot; version in GitHub Codespaces" src="https://github.com/codespaces/badge.svg" /></a></p>
+<p><a href="https://codespaces.new/dynatrace-perfclinics/obslab-llm-observability?ref=ollama-pinecone" target="_blank"><img alt="Open &quot;RAG&quot; version in GitHub Codespaces" src="https://github.com/codespaces/badge.svg" /></a></p>
 <p>Leave the top section blank and provide your values in the <code>Recommended secrets</code> form.</p>
 <p>After the codespaces has started (in a new browser tab), the post creation script should begin. This will install everything and will take a few moments.</p>
 <p>When the script has completed, a success message will briefly be displayed (it is so quick you'll probably miss it) and an empty terminal window will be shown.</p>

visualise-dt/index.html

@@ -684,12 +684,14 @@ <h1 id="visualising-data-in-dynatrace">Visualising Data in Dynatrace<a class="he
 <h2 id="uploading-the-dashboards">Uploading the Dashboards<a class="headerlink" href="#uploading-the-dashboards" title="Permanent link">#</a></h2>
 <p>This demo comes with several prebuilt dashboards. Do the following in Dynatrace.</p>
 <ul>
-<li>Save the contents of <a href="https://github.com/dynatrace-perfclinics/obslab-llm-observability/blob/main/dynatrace/dashboards/openai/Travel-Advisor-Overview.json" target="_blank">dynatrace/dashboards/openai/Travel-Advisor-Overview.json</a> to your computer</li>
+<li>Save the contents of <a href="https://github.com/dynatrace-perfclinics/obslab-llm-observability/blob/main/dynatrace/dashboards/openai/Travel-Advisor-Overview.json" target="_blank">dynatrace/dashboards/openai/Travel-Advisor-Overview.json</a>
+and <a href="https://github.com/Dynatrace/obslab-llm-observability/blob/ollama-pinecone/dynatrace/dashboards/ollama-pinecone/%5BAiObs%5D%20Travel-Advisor-Overview.json">dynatrace/dashboards/ollama-pinecone/[AiObs] Travel-Advisor-Overview.json</a> to your computer</li>
 <li>Press  <code>Ctrl + k</code> and search for <code>dashboards</code> or select the icon from the left toolbar</li>
 <li>Select the <code>Upload</code> button and upload the JSON file.</li>
 </ul>
 <p><img alt="upload button" src="../images/dashboard-upload.png" /></p>
-<p><img alt="dashboard image" src="../images/dashboard.png" /></p>
+<p><img alt="dashboard image" src="../images/dashboard.png" />
+<img alt="dashboard image" src="https://dt-cdn.net/images/ollama-pinecone-dashboard-1713-fa38ba7a33.png" /></p>
 <p>Repeat this process for all the dashboards inside <a href="https://github.com/dynatrace-perfclinics/obslab-llm-observability/blob/main/dynatrace/dashboards" target="_blank">dynatrace/dashboards/*</a></p>
 <h2 id="distributed-traces">Distributed Traces<a class="headerlink" href="#distributed-traces" title="Permanent link">#</a></h2>
 <p>The application emits distributed traces which can be viewed in Dynatrace:</p>