An interview with Mike Kwan, Director of Quantitative Development within Aspect’s Trend Research team

You’re a Director of Quantitative Development within Aspect’s Trend Research team. Please could you tell us about you and your career path so far?

I grew up with an interest in computers and quickly became fascinated with the technical minutiae of low-level programming in assembly language. After studying Computing at university, I went on to join the Technology division at an investment bank where I built iOS and Android mobile applications. Mobile app development was a relatively nascent field at the time and I was fortunate to work on a number of interesting greenfield projects.

Did you always want to work in Quant Dev?

After working on creating mobile UIs for several years, I knew that I wanted to get stuck into further technical challenges but within a deeper domain with more room to grow. I also felt it would be more rewarding to move to a smaller organisation where technology was intertwined with the core business. In hindsight, the move was a really great fit for what I was looking for.

Could you tell us about you and your team’s role? What does it take to make a good QD?

Each investment team at Aspect is composed of an interdisciplinary set of Portfolio Manager/s, Researchers and Quant Developers. Collectively, we create systematic programmes that ingest market data and indicators to determine what position to take for each security in our traded universe. The QD team provisions the technology platform upon which this signal generation, portfolio construction and trade scheduling logic is built.

One unusual characteristic of the platform we maintain is its duality: firstly, to operate as a traditional software system that needs to reliably process vast amounts of data in a performant fashion; and secondly, as an interactive backtest environment that can be flexed to the whims of a continuous research process. The standard engineering practice of defining a compositional framework of blocks is somewhat at odds with the fluidity of research, which requires the ability to rapidly prototype both incremental and structural changes.

The strongest QDs I have come across are excellent communicators and tend to have an almost obsessive drive to scratch beneath the surface to understand how and why things are built the way they are. This allows them to skilfully navigate design trade-offs and adapt pragmatically to the flow of constant change.

What is your favourite part of the role?

Being able to see how the work that we do ties into the business strategy. Projects can be really specific such as modelling a new asset class or squeezing more speed out of a trading strategy. On the other hand, examples of broader projects might be the launch of a new product or investment vehicle. In each case, understanding the mechanics behind the trade lifecycle and how the change impacts the rest of the business is crucial to making the right technology decisions.

What has been the most unexpected thing about your role at Aspect?

Having joined Aspect from a large investment bank, the lack of boundaries between teams was a stark contrast with my experience up to that point. At Aspect, it seemed like people went out of their way to create an accessible and accepting environment and I felt everyone across the business was approachable regardless of their seniority or role.

What exciting developments in the Quant Dev space are you following closely and what new techniques you are exploring?

The growth of the data science industry in the past decade has brought with it huge advances in open-source technology that is applicable to our space. For language-agnostic interoperability, we are particularly excited about the Arrow Flight framework, which allows data streams to be passed between high-performance services.

At a higher level, the complexity and scale of our product range has naturally led us down the path of a cloud compute architecture. In this area, we are pushing towards a fully elastic cloud-native stack. This paradigm shift comes with its own considerations, in some cases demanding a fundamental rethink of data consumption patterns and user workflows.

Are there any blogs or podcasts you follow that you would recommend to someone interested in the industry?

For technology, I like the Lex Fridman Podcast. More specific to our industry, I've listened to a few episodes of Top Traders Unplugged.