diff --git a/Angris/Angris - SEBI (2).pdf b/Angris/Angris - SEBI (2).pdf
new file mode 100644
index 0000000..51f12c2
Binary files /dev/null and b/Angris/Angris - SEBI (2).pdf differ
diff --git a/Angris/readme.md b/Angris/readme.md
new file mode 100644
index 0000000..474e491
--- /dev/null
+++ b/Angris/readme.md
@@ -0,0 +1,38 @@
+#Empowering-Investors-Hackathon
+
+#### Team Name - Angris
+#### Problem Statement - In today's digital age, Indian banks grapple with providing timely, trustworthy educational content on securities markets. Our video generator tool, harnessing Gen AI and Machine Learning, transforms prompts into animated, tailored videos. From broad topics like "Option Trading" to nuanced details, it crafts engaging animations. Banks can now effortlessly educate customers, respond to regulatory changes, and curate financial playlists. We offer a revolution: personalized, engaging, and reliable financial education, all at a click.
+ 
+#### Solution - An application, that can build consistently styled videos just from the prompt. It can crate awareness nad knowledge videos just from the prompt
+
+#### Team Leader Email - vinokrish001@gmail.com
+
+## We are refraining from open-sourcing the code as we are building a product around it  
+
+## Generated final output:
+[![Watch the video](https://img.youtube.com/vi/IyzJQ_OW_Vs/maxresdefault.jpg)](https://www.youtube.com/watch?v=IyzJQ_OW_Vs&ab_channel=KrishnanR)
+
+
+## A Brief of the Prototype:
+ 
+  - We are using fine-tuned GPT with chain of thought prompting to generate the screenplay for the advertisement
+  - The screenplay provides insights which is directed towards stable diffusion to generate the scene environments
+  - The textual screenplay is converted into mp3 using Google Wavenet
+  - Programming and frame manupilation is used with the GPT generated screenplay to generate the final video
+
+## Tech Stack: 
+   - GPT3.5 API 
+   - Stable Diffusion
+   - Google Wavenet
+   - Python Flask Server
+   - MongoDD
+   - Azure VM
+   - Netlify
+
+   
+## What we Learned:
+   - BUILDING RELIABLE GPT APPLICATION IS HARD, WHAT'S HARDER IS BUILDING ONE WHERE THE PROGRAMMING LOGIC IS CONTROLLED BY GPT IS HARDER 
+   - Chain of thought prompting techniques
+   - Integrating responses from different GenAI models  
+   - Building an end to end pipeline for video generation from fundamentals
+   
\ No newline at end of file