diff --git a/bootstraprag/cli.py b/bootstraprag/cli.py
index 065bae8..58bd07f 100644
--- a/bootstraprag/cli.py
+++ b/bootstraprag/cli.py
@@ -22,7 +22,8 @@ def create_zip(project_name):
 @click.option('--observability', type=click.Choice([]), prompt=False)
 def create(project_name, framework, template, observability):
     template_choices = []
-    framework_choices = ['llamaindex', 'langchain', 'standalone-qdrant', 'standalone-evaluations']
+    framework_choices = ['llamaindex', 'langchain', 'standalone-qdrant', 'standalone-evaluations', 'phidata', 'crewai',
+                         'mem0']
     framework = inquirer.select(
         message="Which technology would you like to use?",
         choices=framework_choices
@@ -72,6 +73,19 @@ def create(project_name, framework, template, observability):
             'phoenix-evals',
             'ragas-evals'
         ]
+    elif framework == 'phidata':
+        template_choices = [
+            'agentic-rag'
+        ]
+    elif framework == 'mem0':
+        template_choices = [
+            'personal-ai-assistant-with-memory',
+            'react-agent-with-memory'
+        ]
+    elif framework == 'crewai':
+        template_choices = [
+            'rag-with-crewai-and-llamaindex'
+        ]
     # Use InquirerPy to select template with arrow keys
     template = inquirer.select(
         message="Which template would you like to use?",
diff --git a/bootstraprag/templates/crewai-agents/main.py b/bootstraprag/templates/crewai-agents/main.py
deleted file mode 100644
index 95198d7..0000000
--- a/bootstraprag/templates/crewai-agents/main.py
+++ /dev/null
@@ -1,45 +0,0 @@
-from crewai import Agent, Task, Crew, Process, LLM
-
-llm = LLM(model="ollama/llama3.2:latest", base_url="http://localhost:11434", temperature=0.8, timeout=300)
-
-# Define the Prompt Supervisor agent
-prompt_supervisor = Agent(
-    role='Prompt Supervisor',
-    goal='Ensure all agent prompts are clear, effective, and aligned with users objectives.',
-    backstory=(
-        "As a Prompt Supervisor, you have a keen eye for detail and a deep understanding "
-        "of effective communication strategies. Your mission is to review and refine prompts "
-        "to maximize the performance of AI agents."
-    ),
-    llm=llm
-)
-
-# Define a task for the Prompt Supervisor to review and enhance the Senior Researcher's prompt
-prompt_supervisor_task = Task(
-    description=(
-        "Review the prompt provided to the {topic} Senior prompt supervisor, assessing its clarity, "
-        "effectiveness, and alignment with the project's objectives. Provide constructive feedback "
-        "and suggest improvements to enhance the agent's performance."
-    ),
-    expected_output=(
-        "A detailed evaluation of the original prompt, including specific suggestions for improvement "
-        "and a revised version of the prompt that optimizes clarity and effectiveness along with few shot of examples."
-    ),
-    agent=prompt_supervisor
-)
-
-
-def main():
-    # Forming the crew and kicking off the process
-    crew = Crew(
-        agents=[prompt_supervisor],
-        tasks=[prompt_supervisor_task],
-        process=Process.sequential,
-        verbose=True
-    )
-    result = crew.kickoff(inputs={'topic': 'Get Financial data for 2023'})
-    print(result)
-
-
-if __name__ == "__main__":
-    main()
diff --git a/bootstraprag/templates/crewai-agents/requirements.txt b/bootstraprag/templates/crewai-agents/requirements.txt
deleted file mode 100644
index b9b1d3d..0000000
--- a/bootstraprag/templates/crewai-agents/requirements.txt
+++ /dev/null
@@ -1,2 +0,0 @@
-crewai
-crewai-tools
\ No newline at end of file
diff --git a/bootstraprag/templates/crewai-agents/__init__.py b/bootstraprag/templates/crewai/__init__.py
similarity index 100%
rename from bootstraprag/templates/crewai-agents/__init__.py
rename to bootstraprag/templates/crewai/__init__.py
diff --git a/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/.env b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/.env
new file mode 100644
index 0000000..1dbe14d
--- /dev/null
+++ b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/.env
@@ -0,0 +1,25 @@
+DB_URL='http://localhost:6333'
+DB_API_KEY='th3s3cr3tk3y'
+COLLECTION_NAME='CREWAI_COLLECTION'
+
+OPENAI_API_KEY=''
+OPENAI_EMBED_MODEL='text-embedding-3-small'
+OPENAI_MODEL='gpt-4o'
+
+# use this incase you are prefering to experiment with local models.
+OLLAMA_BASE_URL='http://localhost:11434'
+OLLAMA_LLM_MODEL='qwen2.5:latest'
+OLLAMA_EMBED_MODEL='nomic-embed-text:latest'
+
+# logger can be controlled usiing env
+CRITICAL = 50
+FATAL = 50
+ERROR = 40
+WARNING = 30
+WARN = 30
+INFO = 20
+DEBUG = 10
+NOTSET = 0
+
+LIT_SERVER_WORKERS_PER_DEVICE=2
+LIT_SERVER_PORT=8000
\ No newline at end of file
diff --git a/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/__init__.py b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/api_server.py b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/api_server.py
new file mode 100644
index 0000000..9bebd92
--- /dev/null
+++ b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/api_server.py
@@ -0,0 +1,35 @@
+from abc import ABC
+from dotenv import load_dotenv, find_dotenv
+from crew_agents import rag_crew
+import litserve as ls
+import os
+
+_ = load_dotenv(find_dotenv())
+
+
+class ReactRAGServingAPI(ls.LitAPI, ABC):
+    def __init__(self):
+        self.inputs = {'topic': ''}
+
+    def setup(self, devices):
+        pass
+
+    def decode_request(self, request, **kwargs):
+        return request["query"]
+
+    def predict(self, query: str, **kwargs):
+        try:
+            self.inputs['topic'] = query
+            return rag_crew.kickoff(inputs=self.inputs)
+        except Exception as e:
+            return e.args[0]
+
+    def encode_response(self, output, **kwargs):
+        return {'response': output}
+
+
+if __name__ == '__main__':
+    api = ReactRAGServingAPI()
+    server = ls.LitServer(lit_api=api, api_path='/api/v1/chat-completion',
+                          workers_per_device=int(os.environ.get('LIT_SERVER_WORKERS_PER_DEVICE')))
+    server.run(port=os.environ.get('LIT_SERVER_PORT'))
diff --git a/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/client.py b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/client.py
new file mode 100644
index 0000000..e396db2
--- /dev/null
+++ b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/client.py
@@ -0,0 +1,17 @@
+# Copyright The Lightning AI team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import requests
+
+response = requests.post("http://127.0.0.1:8000/predict", json={"input": 4.0})
+print(f"Status: {response.status_code}\nResponse:\n {response.text}")
diff --git a/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/crew_agents.py b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/crew_agents.py
new file mode 100644
index 0000000..80b1254
--- /dev/null
+++ b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/crew_agents.py
@@ -0,0 +1,84 @@
+from crewai import Agent, Task, Crew, Process, LLM
+from crewai_tools import LlamaIndexTool
+from llama_index.core.tools import FunctionTool
+from llama_index_query_engine import RagQueryEngine
+from typing import Any, Dict
+import nltk
+
+nltk.download('punkt')
+
+rag_query_engine: RagQueryEngine = RagQueryEngine(input_dir='data', show_progress=True)
+
+
+def get_valid_property(data) -> str:
+    # Return the first non-empty, non-"None" value among the keys
+    return (
+        data["description"] if data["description"] not in [None, "None"] else None
+                                                                              or data["text"] if data["text"] not in [
+            None, "None"] else None or
+                               data["content"] if data["content"] not in [None, "None"] else None
+    )
+
+
+def use_query_engine(query: Dict):
+    """Use this function to get answers for mlops
+
+    Args:
+        query (Dict): the user query to search.
+    """
+    query_engine = rag_query_engine.get_query_engine()
+    user_query = get_valid_property(
+        data=query
+    )
+    return query_engine.query(str_or_query_bundle=user_query)
+
+
+query_engine_tool = FunctionTool.from_defaults(
+    use_query_engine,
+    name="mlops tool",
+    description="Use this tool to lookup questions regarding mlops"
+)
+tool = LlamaIndexTool.from_tool(query_engine_tool)
+
+llm = LLM(
+    model="ollama/llama3.2:latest",
+    base_url="http://localhost:11434"
+)
+
+# Initialize Tool from a LlamaIndex Query Engine
+# query_engine = rag_query_engine.get_query_engine()
+# query_tool = LlamaIndexTool.from_query_engine(
+#     query_engine,
+#     name="mlops tool",
+#     description="Use this tool to lookup questions regarding mlops"
+# )
+
+# Create and assign the tools to an agent
+rag_agent = Agent(
+    llm=llm,
+    role='Senior MLops export',
+    goal='Provide up-to-date answer on the user query regarding {topic}',
+    backstory="""As an mlops expert use the tool for fetching the proper context for answering 
+              the user query regarding {topic}""",
+    tools=[tool],
+    max_iter=10,
+    memory=True
+)
+
+rag_task = Task(
+    description="{topic}",
+    expected_output="A summarizing answer for question {topic}.",
+    agent=rag_agent
+)
+
+rag_crew = Crew(
+    agents=[rag_agent],
+    tasks=[rag_task],
+    process=Process.sequential,
+    verbose=True
+)
+
+# Example of using kickoff_async
+# inputs = {'topic': 'what are the challenges of mlops?'}
+# output = rag_crew.kickoff(inputs=inputs)
+# print(output)
diff --git a/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/data/mlops.pdf b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/data/mlops.pdf
new file mode 100644
index 0000000..c8d8170
Binary files /dev/null and b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/data/mlops.pdf differ
diff --git a/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/llama_index_query_engine.py b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/llama_index_query_engine.py
new file mode 100644
index 0000000..bf287b7
--- /dev/null
+++ b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/llama_index_query_engine.py
@@ -0,0 +1,88 @@
+from llama_index.core import (
+    SimpleDirectoryReader,
+    VectorStoreIndex,
+    StorageContext,
+    Settings
+)
+from llama_index.core.base.base_query_engine import BaseQueryEngine
+from llama_index.embeddings.ollama import OllamaEmbedding
+from llama_index.vector_stores.qdrant import QdrantVectorStore
+from llama_index.core.agent import ReActAgent
+from llama_index.llms.ollama import Ollama
+from llama_index.core.base.response.schema import Response, StreamingResponse, AsyncStreamingResponse, PydanticResponse
+from dotenv import load_dotenv, find_dotenv
+from typing import Union
+import qdrant_client
+import logging
+import os
+
+_ = load_dotenv(find_dotenv())
+
+logging.basicConfig(level=int(os.environ['INFO']))
+logger = logging.getLogger(__name__)
+
+
+class RagQueryEngine:
+    RESPONSE_TYPE = Union[
+        Response, StreamingResponse, AsyncStreamingResponse, PydanticResponse
+    ]
+
+    def __init__(self, input_dir: str, similarity_top_k: int = 3, chunk_size: int = 128, chunk_overlap: int = 100,
+                 show_progress: bool = False):
+        self.index_loaded = False
+        self.similarity_top_k = similarity_top_k
+        self.input_dir = input_dir
+        self._index: VectorStoreIndex = None
+        self._engine = None
+        self.agent: ReActAgent = None
+        self.query_engine_tools = []
+        self.show_progress = show_progress
+
+        # use your prefered vector embeddings model
+        logger.info("initializing the OllamaEmbedding")
+        embed_model = OllamaEmbedding(model_name=os.environ['OLLAMA_EMBED_MODEL'],
+                                      base_url=os.environ['OLLAMA_BASE_URL'])
+        # openai embeddings, embedding_model_name="text-embedding-3-large"
+        # embed_model = OpenAIEmbedding(embed_batch_size=10, model=embedding_model_name)
+
+        # use your prefered llm
+        llm = Ollama(model=os.environ['OLLAMA_LLM_MODEL'], base_url=os.environ['OLLAMA_BASE_URL'], request_timeout=600)
+        # llm = OpenAI(model="gpt-4o")
+
+        logger.info("initializing the global settings")
+        Settings.embed_model = embed_model
+        Settings.llm = llm
+        Settings.chunk_size = chunk_size
+        Settings.chunk_overlap = chunk_overlap
+
+        # Create a local Qdrant vector store
+        logger.info("initializing the vector store related objects")
+        self.client: qdrant_client.QdrantClient = qdrant_client.QdrantClient(url=os.environ['DB_URL'],
+                                                                             api_key=os.environ['DB_API_KEY'])
+        self.vector_store = QdrantVectorStore(client=self.client, collection_name=os.environ['COLLECTION_NAME'])
+
+        self._create_index()
+
+    def _create_index(self):
+        if self.client.collection_exists(collection_name=os.environ['COLLECTION_NAME']):
+            try:
+                self._index = VectorStoreIndex.from_vector_store(vector_store=self.vector_store)
+                self.index_loaded = True
+            except Exception as e:
+                self.index_loaded = False
+
+        if not self.index_loaded:
+            # load data
+            _docs = SimpleDirectoryReader(input_dir=self.input_dir).load_data(show_progress=self.show_progress)
+
+            # build and persist index
+            storage_context = StorageContext.from_defaults(vector_store=self.vector_store)
+            logger.info("indexing the docs in VectorStoreIndex")
+            self._index = VectorStoreIndex.from_documents(documents=_docs, storage_context=storage_context,
+                                                          show_progress=self.show_progress)
+
+    def get_query_engine(self) -> BaseQueryEngine:
+
+        logger.info("creating query engine")
+        query_engine = self._index.as_query_engine(similarity_top_k=self.similarity_top_k, llm=Settings.llm)
+        return query_engine
diff --git a/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/requirements.txt b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/requirements.txt
new file mode 100644
index 0000000..b65ab49
--- /dev/null
+++ b/bootstraprag/templates/crewai/rag_with_crewai_and_llamaindex/requirements.txt
@@ -0,0 +1,8 @@
+crewai==0.86.0
+crewai-tools==0.17.0
+llama-index==0.12.5
+llama-index-vector-stores-qdrant==0.4.0
+llama-index-llms-ollama==0.5.0
+llama-index-embeddings-ollama==0.5.0
+qdrant-client==1.12.1
+litserve==0.2.5
\ No newline at end of file
diff --git a/bootstraprag/templates/llamaindex/rag_with_sub_question_query_engine_with_observability/readme.md b/bootstraprag/templates/llamaindex/rag_with_sub_question_query_engine_with_observability/readme.md
index 452a767..4c5106a 100644
--- a/bootstraprag/templates/llamaindex/rag_with_sub_question_query_engine_with_observability/readme.md
+++ b/bootstraprag/templates/llamaindex/rag_with_sub_question_query_engine_with_observability/readme.md
@@ -16,7 +16,7 @@ The Sub-Question Query Engine in LlamaIndex is designed to handle complex querie
 }
 ```
 
-#### How to spin observability
+#### How to spin observability (langfuse)
 - run `docker compose -f docker-compose-langfuse.yml up`
 - launch langfuse in browser `http://localhost:3000`
 - click on `signup`
diff --git a/bootstraprag/templates/mem0/__init__.py b/bootstraprag/templates/mem0/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/.env b/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/.env
new file mode 100644
index 0000000..c0a3e3c
--- /dev/null
+++ b/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/.env
@@ -0,0 +1,3 @@
+OPENAI_API_KEY=sk-proj-
+LIT_SERVER_WORKERS_PER_DEVICE=2
+LIT_SERVER_PORT=8000
\ No newline at end of file
diff --git a/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/__init__.py b/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/api_server.py b/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/api_server.py
new file mode 100644
index 0000000..2139cb3
--- /dev/null
+++ b/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/api_server.py
@@ -0,0 +1,38 @@
+from abc import ABC
+from dotenv import load_dotenv, find_dotenv
+from personal_assistant import PersonalAIAssistant
+import litserve as ls
+import os
+
+_ = load_dotenv(find_dotenv())
+
+
+class ReactRAGServingAPI(ls.LitAPI, ABC):
+    def __init__(self):
+        self.ai_tutor: PersonalAIAssistant = None
+        self.user_id = "pavan_mantha"
+
+    def setup(self, devices):
+        self.ai_tutor: PersonalAIAssistant = PersonalAIAssistant()
+
+    def decode_request(self, request, **kwargs):
+        return request["query"]
+
+    def predict(self, query: str, **kwargs):
+        try:
+            response = self.ai_tutor.ask(question=query, user_id=self.user_id)
+            # you can get the memory value also.
+            # print(self.ai_tutor.get_memories(user_id=self.user_id))
+            return response
+        except Exception as e:
+            return e.args[0]
+
+    def encode_response(self, output, **kwargs):
+        return {'response': output}
+
+
+if __name__ == '__main__':
+    api = ReactRAGServingAPI()
+    server = ls.LitServer(lit_api=api, api_path='/api/v1/chat-completion',
+                          workers_per_device=int(os.environ.get('LIT_SERVER_WORKERS_PER_DEVICE')))
+    server.run(port=os.environ.get('LIT_SERVER_PORT'))
diff --git a/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/client.py b/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/client.py
new file mode 100644
index 0000000..e396db2
--- /dev/null
+++ b/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/client.py
@@ -0,0 +1,17 @@
+# Copyright The Lightning AI team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import requests
+
+response = requests.post("http://127.0.0.1:8000/predict", json={"input": 4.0})
+print(f"Status: {response.status_code}\nResponse:\n {response.text}")
diff --git a/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/personal_assistant.py b/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/personal_assistant.py
new file mode 100644
index 0000000..24145a0
--- /dev/null
+++ b/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/personal_assistant.py
@@ -0,0 +1,79 @@
+from openai import OpenAI
+from mem0 import Memory
+from dotenv import load_dotenv, find_dotenv
+import os
+
+
+load_dotenv(find_dotenv())
+
+# Set the OpenAI API key
+os.environ['OPENAI_API_KEY'] = os.environ.get("OPENAI_API_KEY")
+
+# Initialize the OpenAI client
+client = OpenAI()
+
+
+class PersonalAIAssistant:
+    def __init__(self):
+        """
+        Initialize the PersonalAITutor with memory configuration and OpenAI client.
+        """
+        config = {
+            "vector_store": {
+                "provider": "qdrant",
+                "config": {
+                    "collection_name": "personal_assistant_memory",
+                    "url": "http://localhost:6333",
+                    "api_key": "th3s3cr3tk3y"
+                }
+            },
+        }
+        self.memory = Memory.from_config(config)
+        self.client = client
+        self.app_id = "assistant-app"
+
+    def ask(self, question, user_id=None):
+        """
+        Ask a question to the AI and store the relevant facts in memory
+
+        :param question: The question to ask the AI.
+        :param user_id: Optional user ID to associate with the memory.
+        """
+        # Start a streaming chat completion request to the AI
+        stream = self.client.chat.completions.create(
+            model="gpt-4o-2024-08-06",
+            stream=True,
+            messages=[
+                {"role": "system", "content": "You are a personal AI Assistant."},
+                {"role": "user", "content": question}
+            ]
+        )
+        # Store the question in memory
+        self.memory.add(question, user_id=user_id, metadata={"app_id": self.app_id})
+        response: str = ''
+        # Print the response from the AI in real-time
+        for chunk in stream:
+            if chunk.choices[0].delta.content is not None:
+                print(chunk.choices[0].delta.content, end="")
+                response = response + chunk.choices[0].delta.content
+
+        return response
+
+    def get_memories(self, user_id=None):
+        """
+        Retrieve all memories associated with the given user ID.
+
+        :param user_id: Optional user ID to filter memories.
+        :return: List of memories.
+        """
+        return self.memory.get_all(user_id=user_id)
+
+
+# Instantiate the PersonalAITutor
+# ai_tutor = PersonalAIAssistant()
+
+# Define a user ID
+# user_id = "pavan_mantha"
+
+# Ask a question
+# ai_tutor.ask("What is my first question.", user_id=user_id)
diff --git a/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/requirements.txt b/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/requirements.txt
new file mode 100644
index 0000000..1bb3af6
--- /dev/null
+++ b/bootstraprag/templates/mem0/personal_ai_assistant_with_memory/requirements.txt
@@ -0,0 +1,4 @@
+openai==1.57.4
+mem0ai==0.1.34
+python-dotenv==1.0.1
+litserve==0.2.5
diff --git a/bootstraprag/templates/mem0/react_agent_with_memory/.env b/bootstraprag/templates/mem0/react_agent_with_memory/.env
new file mode 100644
index 0000000..d967bcc
--- /dev/null
+++ b/bootstraprag/templates/mem0/react_agent_with_memory/.env
@@ -0,0 +1,26 @@
+QDRANT_URL='http://localhost:6333'
+QDRANT_API_KEY='th3s3cr3tk3y'
+COLLECTION_NAME='REACT_COLLECTION_WITH_MEM0'
+
+OPENAI_API_KEY=''
+OPENAI_EMBED_MODEL=''
+
+# use this incase you are prefering to experiment with local models.
+OLLAMA_BASE_URL='http://localhost:11434'
+OLLAMA_LLM_MODEL='llama3.2:latest'
+OLLAMA_EMBED_MODEL='nomic-embed-text:latest'
+
+# logger can be controlled usiing env
+CRITICAL = 50
+FATAL = 50
+ERROR = 40
+WARNING = 30
+WARN = 30
+INFO = 20
+DEBUG = 10
+NOTSET = 0
+
+LIT_SERVER_PORT=8000
+LIT_SERVER_WORKERS_PER_DEVICE=4
+
+IS_EVALUATION_NEEDED=true
diff --git a/bootstraprag/templates/mem0/react_agent_with_memory/__init__.py b/bootstraprag/templates/mem0/react_agent_with_memory/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/bootstraprag/templates/mem0/react_agent_with_memory/api_server.py b/bootstraprag/templates/mem0/react_agent_with_memory/api_server.py
new file mode 100644
index 0000000..42d66bd
--- /dev/null
+++ b/bootstraprag/templates/mem0/react_agent_with_memory/api_server.py
@@ -0,0 +1,34 @@
+from abc import ABC
+from dotenv import load_dotenv, find_dotenv
+from react_agent_with_query_engine import ReActWithQueryEngine
+import litserve as ls
+import os
+
+_ = load_dotenv(find_dotenv())
+
+
+class ReactRAGServingAPI(ls.LitAPI, ABC):
+    def __init__(self):
+        self.react_with_engine = None
+
+    def setup(self, devices):
+        self.react_with_engine = ReActWithQueryEngine(input_dir='data', show_progress=True)
+
+    def decode_request(self, request, **kwargs):
+        return request["query"]
+
+    def predict(self, query: str):
+        try:
+            return self.react_with_engine.query(user_query=query)
+        except Exception as e:
+            return e.args[0]
+
+    def encode_response(self, output, **kwargs):
+        return {'response': output}
+
+
+if __name__ == '__main__':
+    api = ReactRAGServingAPI()
+    server = ls.LitServer(lit_api=api, api_path='/api/v1/chat-completion',
+                          workers_per_device=int(os.environ.get('LIT_SERVER_WORKERS_PER_DEVICE')))
+    server.run(port=os.environ.get('LIT_SERVER_PORT'))
diff --git a/bootstraprag/templates/mem0/react_agent_with_memory/client.py b/bootstraprag/templates/mem0/react_agent_with_memory/client.py
new file mode 100644
index 0000000..12b71e5
--- /dev/null
+++ b/bootstraprag/templates/mem0/react_agent_with_memory/client.py
@@ -0,0 +1,18 @@
+
+# Copyright The Lightning AI team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import requests
+
+response = requests.post("http://127.0.0.1:8000/predict", json={"input": 4.0})
+print(f"Status: {response.status_code}\nResponse:\n {response.text}")
diff --git a/bootstraprag/templates/mem0/react_agent_with_memory/data/mlops.pdf b/bootstraprag/templates/mem0/react_agent_with_memory/data/mlops.pdf
new file mode 100644
index 0000000..c8d8170
Binary files /dev/null and b/bootstraprag/templates/mem0/react_agent_with_memory/data/mlops.pdf differ
diff --git a/bootstraprag/templates/mem0/react_agent_with_memory/mem0_configs.py b/bootstraprag/templates/mem0/react_agent_with_memory/mem0_configs.py
new file mode 100644
index 0000000..bc227db
--- /dev/null
+++ b/bootstraprag/templates/mem0/react_agent_with_memory/mem0_configs.py
@@ -0,0 +1,36 @@
+import os
+import uuid
+from dotenv import load_dotenv, find_dotenv
+
+load_dotenv(find_dotenv())
+
+context = {
+    "user_id": "pavan_mantha",
+    "agent_id": "react_agent",
+    "run_id": str(uuid.uuid4()),
+}
+
+config = {
+    "vector_store": {
+        "provider": "qdrant",
+        "config": {
+            "collection_name": os.environ.get('COLLECTION_NAME'),
+            "url": os.environ.get('QDRANT_URL'),
+            "api_key": os.environ.get('QDRANT_API_KEY'),
+            "embedding_model_dims": 768,  # Change this according to your local model's dimensions
+        },
+    },
+    "llm": {
+        "provider": "ollama",
+        "config": {
+            "model": "llama3.2:latest",
+            "temperature": 0.2,
+            "max_tokens": 1500,
+        },
+    },
+    "embedder": {
+        "provider": "ollama",
+        "config": {"model": "nomic-embed-text:latest"},
+    },
+    "version": "v1.1",
+}
\ No newline at end of file
diff --git a/bootstraprag/templates/mem0/react_agent_with_memory/react_agent_with_query_engine.py b/bootstraprag/templates/mem0/react_agent_with_memory/react_agent_with_query_engine.py
new file mode 100644
index 0000000..e449f6a
--- /dev/null
+++ b/bootstraprag/templates/mem0/react_agent_with_memory/react_agent_with_query_engine.py
@@ -0,0 +1,134 @@
+from llama_index.core import (
+    SimpleDirectoryReader,
+    VectorStoreIndex,
+    StorageContext,
+    Settings
+)
+from llama_index.core.tools import QueryEngineTool, ToolMetadata
+from llama_index.embeddings.ollama import OllamaEmbedding
+from llama_index.vector_stores.qdrant import QdrantVectorStore
+from llama_index.core.agent import ReActAgent
+from llama_index.llms.ollama import Ollama
+from llama_index.core.base.response.schema import Response, StreamingResponse, AsyncStreamingResponse, PydanticResponse
+from dotenv import load_dotenv, find_dotenv
+from typing import Union
+from mem0_configs import config, context
+from llama_index.memory.mem0 import Mem0Memory
+import qdrant_client
+import logging
+import os
+
+_ = load_dotenv(find_dotenv())
+
+logging.basicConfig(level=int(os.environ['INFO']))
+logger = logging.getLogger(__name__)
+
+
+class ReActWithQueryEngine:
+    RESPONSE_TYPE = Union[
+        Response, StreamingResponse, AsyncStreamingResponse, PydanticResponse
+    ]
+
+    def __init__(self, input_dir: str, similarity_top_k: int = 3, chunk_size: int = 128, chunk_overlap: int = 100,
+                 show_progress: bool = False, no_of_iterations: int = 5, required_exts: list[str] = ['.pdf', '.txt']):
+        self.index_loaded = False
+        self.similarity_top_k = similarity_top_k
+        self.input_dir = input_dir
+        self._index = None
+        self._engine = None
+        self.agent: ReActAgent = None
+        self.query_engine_tools = []
+        self.show_progress = show_progress
+        self.no_of_iterations = no_of_iterations
+        self.required_exts = required_exts
+
+        # the mem0 configs for llamaindex
+        self.memory = Mem0Memory.from_config(
+            context=context,
+            config=config,
+            search_msg_limit=10  # default is 5
+        )
+
+        # use your prefered vector embeddings model
+        logger.info("initializing the OllamaEmbedding")
+        embed_model = OllamaEmbedding(model_name=os.environ['OLLAMA_EMBED_MODEL'],
+                                      base_url=os.environ['OLLAMA_BASE_URL'])
+        # openai embeddings, embedding_model_name="text-embedding-3-large"
+        # embed_model = OpenAIEmbedding(embed_batch_size=10, model=embedding_model_name)
+
+        # use your prefered llm
+        llm = Ollama(model=os.environ['OLLAMA_LLM_MODEL'], base_url=os.environ['OLLAMA_BASE_URL'], request_timeout=600)
+        # llm = OpenAI(model="gpt-4o")
+
+        logger.info("initializing the global settings")
+        Settings.embed_model = embed_model
+        Settings.llm = llm
+        Settings.chunk_size = chunk_size
+        Settings.chunk_overlap = chunk_overlap
+
+        # Create a local Qdrant vector store
+        logger.info("initializing the vector store related objects")
+        self.client: qdrant_client.QdrantClient = qdrant_client.QdrantClient(url=os.environ['QDRANT_URL'],
+                                                                             api_key=os.environ['QDRANT_API_KEY'])
+        self.vector_store = QdrantVectorStore(client=self.client, collection_name=os.environ['COLLECTION_NAME'])
+        self._load_data_and_create_engine()
+
+    def _load_data_and_create_engine(self):
+        if self.client.collection_exists(collection_name=os.environ['COLLECTION_NAME']):
+            try:
+                self._index = VectorStoreIndex.from_vector_store(vector_store=self.vector_store)
+                self.index_loaded = True
+            except Exception as e:
+                self.index_loaded = False
+
+        if not self.index_loaded:
+            # load data
+            _docs = (SimpleDirectoryReader(input_dir=self.input_dir, required_exts=self.required_exts)
+                     .load_data(show_progress=self.show_progress))
+
+            # build and persist index
+            storage_context = StorageContext.from_defaults(vector_store=self.vector_store)
+            logger.info("indexing the docs in VectorStoreIndex")
+            self._index = VectorStoreIndex.from_documents(documents=_docs, storage_context=storage_context,
+                                                          show_progress=self.show_progress)
+
+        self._engine = self._index.as_query_engine(similarity_top_k=self.similarity_top_k)
+        self._create_query_engine_tools()
+
+    def _create_query_engine_tools(self):
+        # can have more than one as per the requirement
+        self.query_engine_tools.append(
+            QueryEngineTool(
+                query_engine=self._engine,
+                metadata=ToolMetadata(
+                    name="react_tool_engine",  # change this accordingly
+                    description=(
+                        "Provides information about user query based on the information that you have. "
+                        "Use a detailed plain text question as input to the tool."
+                    ),
+                ),
+            )
+        )
+        self._create_react_agent()
+
+    def _create_react_agent(self):
+        # [Optional] Add Context
+        # context = """\
+        # You are a stock market sorcerer who is an expert on the companies Lyft and Uber.\
+        #     You will answer questions about Uber and Lyft as in the persona of a sorcerer \
+        #     and veteran stock market investor.
+        # """
+        self.agent = ReActAgent.from_tools(
+            self.query_engine_tools,
+            llm=Settings.llm,
+            verbose=True,
+            # context=context
+            max_iterations=self.no_of_iterations,
+            memory=self.memory
+        )
+
+    def query(self, user_query: str) -> RESPONSE_TYPE:
+        try:
+            return self.agent.query(str_or_query_bundle=user_query)
+        except Exception as e:
+            logger.error(f'Error while generating response: {e}')
diff --git a/bootstraprag/templates/mem0/react_agent_with_memory/requirements.txt b/bootstraprag/templates/mem0/react_agent_with_memory/requirements.txt
new file mode 100644
index 0000000..d2deece
--- /dev/null
+++ b/bootstraprag/templates/mem0/react_agent_with_memory/requirements.txt
@@ -0,0 +1,10 @@
+python-dotenv==1.0.1
+llama-index==0.12.5
+llama-index-llms-ollama==0.5.0
+llama-index-embeddings-ollama==0.5.0
+llama-index-vector-stores-qdrant==0.4.0
+qdrant-client==1.12.1
+llama-index-memory-mem0==0.2.1
+mem0ai==0.1.34
+litserve==0.2.5
+
diff --git a/bootstraprag/templates/phidata/__init__.py b/bootstraprag/templates/phidata/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/bootstraprag/templates/phidata/agentic_rag/.env b/bootstraprag/templates/phidata/agentic_rag/.env
new file mode 100644
index 0000000..adc23b8
--- /dev/null
+++ b/bootstraprag/templates/phidata/agentic_rag/.env
@@ -0,0 +1,7 @@
+OPENAI_API_KEY=sk-proj-
+QDRANT_API_KEY=th3s3cr3tk3y
+QDRANT_URL=http://localhost:6333
+collection_name=thai-phidata-index
+
+LIT_SERVER_WORKERS_PER_DEVICE=2
+LIT_SERVER_PORT=8000
\ No newline at end of file
diff --git a/bootstraprag/templates/phidata/agentic_rag/__init__.py b/bootstraprag/templates/phidata/agentic_rag/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/bootstraprag/templates/phidata/agentic_rag/agentic_rag.py b/bootstraprag/templates/phidata/agentic_rag/agentic_rag.py
new file mode 100644
index 0000000..3e5131e
--- /dev/null
+++ b/bootstraprag/templates/phidata/agentic_rag/agentic_rag.py
@@ -0,0 +1,71 @@
+import os
+import typer
+import qdrant_client
+from typing import Optional
+from rich.prompt import Prompt
+
+from phi.agent import Agent
+from phi.model.ollama import Ollama
+from phi.embedder.ollama import OllamaEmbedder
+from phi.knowledge.pdf import PDFKnowledgeBase, PDFUrlKnowledgeBase
+from phi.vectordb.qdrant import Qdrant
+from dotenv import load_dotenv, find_dotenv
+
+load_dotenv(find_dotenv())
+
+api_key = os.getenv("QDRANT_API_KEY")
+qdrant_url = os.getenv("QDRANT_URL")
+collection_name = os.getenv("collection_name")
+
+qdrantClient = qdrant_client.QdrantClient(url=qdrant_url, api_key=api_key)
+
+vector_db = Qdrant(
+    collection=collection_name,
+    embedder=OllamaEmbedder(model="nomic-embed-text:latest", dimensions=768, host="http://localhost:11434"),
+    url=qdrant_url,
+    api_key=api_key,
+)
+
+knowledge_base = PDFKnowledgeBase(
+    path="data",
+    vector_db=vector_db,
+)
+
+# Comment out after first run
+if not qdrantClient.collection_exists(collection_name=collection_name):
+    knowledge_base.load(recreate=True, upsert=True, skip_existing=True)
+
+
+def qdrant_agent(user: str = "user"):
+    run_id: Optional[str] = None
+
+    agent = Agent(
+        run_id=run_id,
+        model=Ollama(id="llama3.2:latest", host="http://localhost:11434"),
+        user_id=user,
+        knowledge=knowledge_base,
+        tool_calls=True,
+        use_tools=True,
+        show_tool_calls=True,
+        debug_mode=True,
+    )
+
+    if run_id is None:
+        run_id = agent.run_id
+        print(f"Started Run: {run_id}\n")
+    else:
+        print(f"Continuing Run: {run_id}\n")
+
+    return agent
+
+
+# if __name__ == "__main__":
+#     _agent = qdrant_agent('pavan_mantha')
+#     user_id = 'pavan_mantha' # change the user id accordingly
+#     while True:
+#         message = Prompt.ask(f"[bold] :sunglasses: {user_id} [/bold]")
+#         if message in ("exit", "bye"):
+#             break
+#         # _agent.print_response(message)
+#         resp = _agent.run(message=message)
+#         print(resp)
diff --git a/bootstraprag/templates/phidata/agentic_rag/api_server.py b/bootstraprag/templates/phidata/agentic_rag/api_server.py
new file mode 100644
index 0000000..995bcc1
--- /dev/null
+++ b/bootstraprag/templates/phidata/agentic_rag/api_server.py
@@ -0,0 +1,38 @@
+from abc import ABC
+from dotenv import load_dotenv, find_dotenv
+from agentic_rag import qdrant_agent
+import litserve as ls
+import os
+
+_ = load_dotenv(find_dotenv())
+
+
+class ReactRAGServingAPI(ls.LitAPI, ABC):
+    def __init__(self):
+        self.ai_agent = None
+        self.user_id = "pavan_mantha"
+
+    def setup(self, devices):
+        self.ai_agent = qdrant_agent(user='pavan_mantha')
+
+    def decode_request(self, request, **kwargs):
+        return request["query"]
+
+    def predict(self, query: str, **kwargs):
+        try:
+            response = self.ai_agent.run(message=query)
+            # you can get the memory value also.
+            # print(self.ai_tutor.get_memories(user_id=self.user_id))
+            return response
+        except Exception as e:
+            return e.args[0]
+
+    def encode_response(self, output, **kwargs):
+        return {'response': output}
+
+
+if __name__ == '__main__':
+    api = ReactRAGServingAPI()
+    server = ls.LitServer(lit_api=api, api_path='/api/v1/chat-completion',
+                          workers_per_device=int(os.environ.get('LIT_SERVER_WORKERS_PER_DEVICE')))
+    server.run(port=os.environ.get('LIT_SERVER_PORT'))
diff --git a/bootstraprag/templates/phidata/agentic_rag/client.py b/bootstraprag/templates/phidata/agentic_rag/client.py
new file mode 100644
index 0000000..e396db2
--- /dev/null
+++ b/bootstraprag/templates/phidata/agentic_rag/client.py
@@ -0,0 +1,17 @@
+# Copyright The Lightning AI team.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import requests
+
+response = requests.post("http://127.0.0.1:8000/predict", json={"input": 4.0})
+print(f"Status: {response.status_code}\nResponse:\n {response.text}")
diff --git a/bootstraprag/templates/phidata/agentic_rag/data/mlops.pdf b/bootstraprag/templates/phidata/agentic_rag/data/mlops.pdf
new file mode 100644
index 0000000..c8d8170
Binary files /dev/null and b/bootstraprag/templates/phidata/agentic_rag/data/mlops.pdf differ
diff --git a/bootstraprag/templates/phidata/agentic_rag/requirements.txt b/bootstraprag/templates/phidata/agentic_rag/requirements.txt
new file mode 100644
index 0000000..135de53
--- /dev/null
+++ b/bootstraprag/templates/phidata/agentic_rag/requirements.txt
@@ -0,0 +1,4 @@
+phidata==2.7.2
+python-dotenv==1.0.1
+qdrant-client==1.12.1
+litserve==0.2.5
\ No newline at end of file