Updated the AI section

index.html

@@ -176,7 +176,14 @@
         -->
         <div id="AI-Studies" class="tab-content">
             <h4>AI Studies</h4>
-            <p>AI experimental articles Coming soon..</p>
+            <p>
+              AI experimental articles
+              To run AI models locally, we need a runtime engine (inference), decent machine, preferably with few gpu's to support and reasonable performance.
+              We used the following, Ollama, llama.cpp, mlx_lm etc to run few models like llama, gemma, phi to obtain decent performance.
+              We used openlit, pheonix as llm observability tools to evaluate the model performances.
+              We used nomic-embed, all-minilm like embedding models with chroma_db like vector database to test the RAG concepts
+              Detailed experiments and some interesting observations will be published soon ..
+            </p>
         </div>
         <!--
         <div id="Devotional" class="tab-content">