Title

AWS re:Invent 2023 - Modernizing data architecture with data products in asset management (BIZ110)

Summary

Saket, CEO of Nexla, introduces the session and is joined by Chitra Hotta from Oak Tree and Daryl Cherry from Clearwater Analytics.
Chitra discusses her role in building data engineering pipelines, data lakes, and warehouses for analytics and reporting at Oak Tree.
Daryl emphasizes the importance of data in Clearwater Analytics, dealing with large volumes of data from various sources for investment analytics and compliance.
The panel discusses the challenges of handling unstructured data in asset management, such as PDFs, text files, and Excel documents.
Data architecture must be simple, flexible, and capable of processing both structured and unstructured data, with data products playing a critical role in enabling easy data transformation and reducing the need for data engineers.
The use of generative AI and large language models (LLMs) is highlighted, particularly in dealing with voice and unstructured data for insights and summarization.
Practical use cases include transcribing earnings calls for sentiment analysis and summarizing quarterly reports for investment decisions.
AWS cloud-native services like AWS Transcribe, Textract, Glue, EventBridge, Lambda, and others are leveraged for scaling and data processing.
The role of investors is changing with technology, allowing for more informed and quicker decision-making.
Questions from the audience focus on solutions for private credit and the use of public data for investment decisions.

The financial services asset management sector is increasingly relying on a mix of structured and unstructured data, with unstructured data posing significant challenges due to its varied formats.
Data products are essential for simplifying the transformation of data and shortening the time to market for data platforms, highlighting the importance of data architecture in asset management.
Generative AI and LLMs are becoming integral to data architecture, particularly in parsing and summarizing unstructured data like PDFs and voice recordings, which can then be used for further analysis and decision-making.
AWS cloud-native services are critical for handling the scale and variety of data in asset management, with specific services like AWS Transcribe, Textract, Glue, EventBridge, and Lambda being mentioned for their scalability and data processing capabilities.
The advancement of technology, particularly generative AI, is democratizing access to complex information, enabling less experienced users to become more proficient and make informed decisions more rapidly.
The session underscores the ongoing evolution of data architecture in asset management, driven by the need to integrate diverse data sources and leverage AI for actionable insights.