Bin Ren, founder of SigTech, discusses the design of a state-of-the-art backtesting engine and the integration of large language models into the process. Topics include modular design, strategy definition, coupling engine with data, challenges of incorporating alternative and unstructured data, understanding query and making API calls, implications of large language models in programming languages, and contrasting cultural reactions to AI.
Flexibility and control in expressing trade ideas and structuring trades within a state-of-the-art backtesting engine are crucial.
Integrating large language models (LLMs) into the backtesting process requires fine-tuning and efficient management of latency.
The integration of alternative and unstructured data into the backtesting engine presents challenges in quality control, data mapping, and pipeline design.
Deep dives
Designing a State-of-the-Art Backtesting Engine
Ben Ren, founder of SIGTech, discusses the design and development of a state-of-the-art backtesting engine. This includes topics such as monolithic versus modular design, the relationship between the engine and data, and the distinction between strategy definition and backtesting. Ren emphasizes the importance of flexibility and control in expressing trade ideas and structuring trades within the engine.
Integrating Large Language Models into Quantitative Research
Ben Ren explains how SIGTech has incorporated large language models (LLMs) into their process. The discussion covers the technical implications of integrating LLMs, as well as the philosophical view of the evolving role of a quant researcher as AI becomes more prevalent. Ren highlights the need for fine-tuning LLMs and the challenge of managing latency when using multiple APIs in the interaction between the models and the backtesting engine.
Unleashing the Power of Alternative and Unstructured Data
Ben Ren delves into the integration of alternative and unstructured data into the design of the backtesting engine. He talks about the challenges of quality control and mapping alternative time series data to financial instruments. Ren also discusses the complexities of building a data pipeline for unstructured data and the considerations to minimize latency and ensure reliable output when using large language models to process this type of data.
Integration of back testing engine technology with large language models
The podcast discusses how the integration of back testing engine technology with large language models is a significant shift in the business of SigTech. This integration aims to improve the user experience by providing access to high-quality data, reducing the need for extensive programming knowledge, and leveraging AI models as potential customers. The use of large language models, such as GPT-4, allows users to interact with AI through natural language and enhance their problem-solving abilities.
Designing APIs for large language models and the future of quant researchers
The podcast explores the design philosophies for APIs used by large language models, highlighting two major limitations: lack of access to updated data and the need for tools to complete tasks in different domains. The ability of large language models to write code on the fly and generate customized solutions offers a new level of flexibility and creativity. This shift may lead to a change in the role of quant researchers, focusing more on asking the right questions rather than implementing solutions. The democratization of access to data and AI models is expected to increase market participation and lead to a more efficient financial market.
In this episode I speak with Bin Ren, founder of SigTech, a financial technology platform providing quantitative researchers with access to a state-of-the-art analysis engine.
This conversation is really broken into two parts. In the first half, we discuss Bin’s views on designing and developing a state-of-the-art backtesting engine. This includes concepts around monolithic versus modular design, how tightly coupled the engine and data should be, and the blurred line between where a strategy definition ends and the backtest engine begins.
In the second half of the conversation we discuss the significant pivot SigTech has undergone this year to incorporate large language models into its process. Or, perhaps more accurately, allow large language models to be a client to its data and services. Here Bin shares his thoughts on both the technical ramifications of integrating with LLMs as well as his philosophical views as to how the role of a quant researcher will change over time as AI becomes more prevalent.
I hope you enjoy my conversation with Bin Ren.
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
