
LANGCHAIN — Does Morningstar Intelligence Engine Provide Personalized Investment Insights for Analysts?
In theory, there is no difference between theory and practice. But, in practice, there is. — Jan L.A. van de Snepscheut
Morningstar Intelligence Engine (MIE) is a tool designed to provide personalized investment insights to analysts and financial professionals. MIE leverages Generative AI to make Morningstar’s extensive research database more accessible and immediately useful to a wider range of users. Morningstar developed a chatbot, Mo, that allows customers to query their research database using natural language and receive concise yet nuanced insights in seconds. Additionally, Morningstar enables their customers to build white-labeled chatbots and other AI tools of their own through the MIE platform.
Several members of the Morningstar team were familiar with LangChain through their contributions to the open-source repository and chose LangChain to build a production-ready application due to its robust framework and support for initiating LLM-powered application development. The integration of various essential concepts from LangChain, including prompt templates, RAG-based approach, and the ReAct framework, helped expedite the development process.
Here’s an example demonstrating how to use LangChain to initiate an LLM-powered application:
from langchain import LLMApplication
# Initialize the LLM-powered application
app = LLMApplication()
# Query the Morningstar research database using natural language
query = "What are the latest insights on technology stocks?"
insights = app.query_database(query)
print(insights)Furthermore, the LangChain framework provides support for edge cases like retries, enhancing the robustness of the application. LangChain also offers wide-ranging support across various vector databases, improving versatility and applicability.
Here’s an example showcasing how LangChain supports edge cases like retries:
from langchain import LLMApplication
# Initialize the LLM-powered application with retry support
app = LLMApplication(retries=3)
# Query the Morningstar research database using natural language
query = "What are the latest insights on technology stocks?"
insights = app.query_database(query)
print(insights)Additionally, LangChain introduces critical cognitive architectures that facilitate a better grasp of generative AI, enriching the team’s understanding of the evolving technology.
In conclusion, Morningstar Intelligence Engine leverages LangChain to build a production-ready application, enabling the delivery of cutting-edge insights and tools to help financial professionals make better investment decisions.
