diff --git a/backend/main.py b/backend/main.py
index 987203d..f2ddcaa 100644
--- a/backend/main.py
+++ b/backend/main.py
@@ -12,9 +12,11 @@
 from jose import JWTError, jwt
 from langchain_core.messages.ai import AIMessage, AIMessageChunk
 
+from backend.config import RagConfig
 from backend.database import Database
 from backend.logger import get_logger
 from backend.model import Message
+from backend.rag_components.chat_message_history import get_conversation_buffer_memory
 from backend.rag_components.rag import RAG
 from backend.user_management import (
     ALGORITHM,
@@ -89,7 +91,13 @@ async def chat_prompt(message: Message, current_user: User = Depends(get_current
     }
     rag = RAG(config=Path(__file__).parent / "config.yaml", logger=logger, context=context)
     response = rag.generate_response(message)
-    return StreamingResponse(stream_response(message.chat_id, response), media_type="text/event-stream")
+    response_stream = stream_response(
+        rag=rag,
+        chat_id=message.chat_id,
+        question=message.content,
+        response=response
+    )
+    return StreamingResponse(response_stream, media_type="text/event-stream")
 
 
 @app.post("/chat/regenerate")
@@ -130,7 +138,7 @@ async def chat(chat_id: str, current_user: User = Depends(get_current_user)) ->
     return {"chat_id": chat_id, "messages": [message.model_dump() for message in messages]}
 
 
-async def stream_response(chat_id: str, response):
+async def stream_response(rag: RAG, chat_id: str, question, response):
     full_response = ""
     response_id = str(uuid4())
     try:
@@ -154,6 +162,7 @@ async def stream_response(chat_id: str, response):
         yield full_response.encode("utf-8")
     finally:
         await log_response_to_db(chat_id, full_response)
+        await memorize_response(rag.config, chat_id, question, full_response)
 
 async def log_response_to_db(chat_id: str, full_response: str):
     response_id = str(uuid4())
@@ -163,6 +172,10 @@ async def log_response_to_db(chat_id: str, full_response: str):
             (response_id, datetime.now().isoformat(), chat_id, "assistant", full_response),
         )
 
+async def memorize_response(rag_config: RagConfig, chat_id: str, question: str, answer: str):
+    memory = get_conversation_buffer_memory(rag_config, chat_id)
+    memory.save_context({"question": question}, {"answer": answer})
+
 
 ############################################
 ###               Feedback               ###
diff --git a/docs/recipe_vector_stores_configs.md b/docs/recipe_vector_stores_configs.md
index e58c321..bb79a31 100644
--- a/docs/recipe_vector_stores_configs.md
+++ b/docs/recipe_vector_stores_configs.md
@@ -5,7 +5,7 @@ As we need a backend SQL database to store conversation history and other info,
 [See the recipes for database configs here](recipe_databases_configs.md)
 
 ```shell
-pip install pgvector
+pip install psycopg2-binary pgvector
 ```
 
 ```yaml