Interrogation to Mitigate Downside Risk

Pre-Acquisition / Credit Diagnostics

$2 BI

Nulla vitae elit libero

$4.3 MI

Fusce dapibus tellus

73%

Ipsum libero

Goal

Public and private credit investor manager aimed to streamline its credit risk process with an interactive investment knowledge chatbot to assist analysts in extracting and interrogating insights.

Approach

Mapped the investment process to define requirements for decision thresholds, UI, workflow, response time, and data management.

• Built a conversational chat capability with a vector database for a given set of files, supplemented by a chat history to support follow-up questions.

• Extended the functionality with a PDF API to load temporary files for Q&A to support ad-hoc or independent topic research.

• Developed a proprietary task agent with the ability to generate, execute, and adapt a transparent set of actions to answer complex questions.

• Enhanced with a Chain-of-Thought-based agent to search and retrieve deeper insights from online pages.

• Deployed the co-pilot with a containerized UI, API, and DB in a mono-repo structure.

• Utilized a Redis database with an in-memory data store for vector and text-based searches using HashMaps, along with a FastAPI cache for fast responses from static API endpoints.

Interventions

Delivered a chat-based co-pilot workflow with the following capabilities:

• Summarization and key point extraction from multi-modal data sources and documentation.

• Conversational interface with the ability to rephrase a question and create a plan to improve accuracy.

• Planning agent capable of creating plans leveraging multiple tools, steps, and sources, with self-correction when needed.

Additionally:

• Up-skilled the internal data science team on technical best practices through day-to-day collaboration and on-the-job coaching.