diff --git a/src/constants/MarkdownFiles/authors/mebin-thattil.md b/src/constants/MarkdownFiles/authors/mebin-thattil.md
index 5f17e7c8..0723865a 100644
--- a/src/constants/MarkdownFiles/authors/mebin-thattil.md
+++ b/src/constants/MarkdownFiles/authors/mebin-thattil.md
@@ -4,7 +4,7 @@ slug: "mebin-thattil"
 title: "GSoC'25 Contributor"
 organization: "SugarLabs"
 description: "GSoC'25 Contributor at SugarLabs"
-avatar: "https://mebin.shop/mebin-380.webp"
+avatar: "https://cdn.mebin.in/mebin-380.webp"
 ---
 
 <!--markdownlint-disable-->
diff --git a/src/constants/MarkdownFiles/posts/2025-06-14-gsoc-25-MebinThattil-week02.md b/src/constants/MarkdownFiles/posts/2025-06-14-gsoc-25-MebinThattil-week02.md
index 9f6b273a..2a53dbc0 100644
--- a/src/constants/MarkdownFiles/posts/2025-06-14-gsoc-25-MebinThattil-week02.md
+++ b/src/constants/MarkdownFiles/posts/2025-06-14-gsoc-25-MebinThattil-week02.md
@@ -104,7 +104,7 @@ image: "assets/Images/GSOCxSpeak.webp"
 
 ## Sample model output with stop condition
 
-![sample model output](https://mebin.shop/Mebin-test-FT-model-tesponses.webp)
+![sample model output](https://cdn.mebin.in/Mebin-test-FT-model-tesponses.webp)
 
 ---
 
diff --git a/src/constants/MarkdownFiles/posts/2025-07-05-gsoc-25-MebinThattil-week05.md b/src/constants/MarkdownFiles/posts/2025-07-05-gsoc-25-MebinThattil-week05.md
index 4eb9f2c3..45f04292 100644
--- a/src/constants/MarkdownFiles/posts/2025-07-05-gsoc-25-MebinThattil-week05.md
+++ b/src/constants/MarkdownFiles/posts/2025-07-05-gsoc-25-MebinThattil-week05.md
@@ -46,7 +46,7 @@ This week, I tested out different voices of Kokoro in two different ways:
     - The primary reason this was built as a web app is so that we can get kids to test this out and having things as a web app makes it easier. It's cruical for us to get real world feedback before proceeding with the implementation.
     - This web app allows you to try out a plethora of different voices and also mix and match different voices to create basically infinite combinations. It's truly amazing the kind of voices you can create with this.
 
-    ![UI of web app](https://mebin.shop/Kokoro-Streamlit-UI.webp)
+    ![UI of web app](https://cdn.mebin.in/Kokoro-Streamlit-UI.webp)
 
     - It's a container app, meaning both the frontend (Streamlit) and backend (Kokoro - FastAPI) run as separate Docker containers hosted on Azure.  
     - The [Kokoro - FastAPI](https://github.com/mebinthattil/Kokoro-FastAPI) exposes an OpenAI-compatible API to generate audio.