“Hedge funds’ changing needs mean we must prioritise agility”
Abhinandan Kumar, Vice President Sales for APAC at KX, spoke to the Hedge Funds Club’s Stefan Nilsson about how the firm helps its hedge fund clients to gain an edge with AI. KX powers the time-aware data-driven decisions that enable fast-moving organisations to realise the full potential of their AI investments and outpace competitors.
What are the key challenges for hedge funds when it comes to data analytics?
Hedge funds are constantly balancing a need for new and more data with the challenge of finding technology that is powerful enough to integrate vast, fast-moving datasets in real time. In conversations with customers, I am often asked for recommendations on new, impactful data sources a company could use to inform their analytics and find alpha. With this constant desire to incorporate new data sources, hedge funds are also increasingly looking for a single platform that goes beyond just data ingestion or capture. They want platforms that can also transform the data and publish it in real time, enabling faster, more actionable insights across the entire trading workflow. Unfortunately, most current solutions are very fragmented; for example, one tool is available for ingesting data, another for capturing it and a separate one for real-time analytics and publishing.
How can hedge funds build a unified, real-time data ecosystem that is fit for a trading world where milliseconds matter?
To build a unified, real-time ecosystem, hedge funds should consolidate disparate data sources (structured and unstructured) into a single platform capable of processing both real-time and historical data with ultra-low latency. Leveraging technologies like in-memory databases, event-driven architectures and efficient time-series data handling is important, as is data normalisation, governance and integration with advanced analytics to enhance decision-making. A flexible, scalable infrastructure that seamlessly connects cloud and on-prem environments ensures agility. By prioritising data timeliness, not just raw speed, funds can maintain a decisive trading edge.
AI has been hyped and overhyped. Now, as things have calmed down and we can see the real AI opportunities, how does KX provide AI solutions for hedge funds?
At KX, we unify capabilities for ingesting and capturing data, real-time analytics and publishing in a single, integrated platform. We offer kdb+ as the database to store all the data, KX Insights Enterprise and the Q language for data transformation, KX Dashboards for publishing and visualising and KX SQL for querying. Additionally, PyKX integrates kdb+ with Python, which is the primary language of choice for coders. We’re the only vendor that delivers this complete, end-to-end solution seamlessly. Additionally, we deliver a generative AI service to customers, called Temporal Search, which uses real-time data to test patterns or trends against a hedge fund’s historical data to identify anomalies or forecast future events. This is critical because, if an LLM is tuned without consideration of time, it may inadvertently learn future patterns through unbounded financial data, ingest general web information containing predictions or leaked information, or experience subtle contamination of supposedly time-bounded training datasets.
Alternative data is something many hedge funds are increasingly using. What are the trends you are seeing there?
Hedge funds increasingly seek to harness alternative data, from satellite imagery to social media sentiment and ESG metrics, to uncover hidden market signals. The trend is toward integrating these diverse datasets directly into core analytics platforms in real time. There’s also a growing emphasis on data provenance, quality and regulatory compliance, ensuring this data can be trusted for decision-making.
Are you facing challenges in product development? As your hedge fund clients’ needs evolve so fast, can you keep up with the demands?
Our hedge fund clients’ changing needs mean that we must prioritise agility in product development. We are able to iterate quickly, deliver scalable solutions and enhance interoperability to meet bespoke analytics needs. To keep pace, we are closely engaging with clients, anticipating emerging requirements, like alternative data integration or cloud-native analytics, and ensuring our platform adapts rapidly.
Do you see any major differences in what hedge funds in Asia need and want versus the demand you have from hedge funds elsewhere?
I wouldn’t say there’s a major difference between what hedge funds in Asia need compared to other regions, but there are some distinctly different priorities. In Asia, I’ve noticed a stronger interest in how generative AI can directly help them generate higher revenue and deliver better alpha, as opposed to just accessing real-time data or running analytics. Many CXOs and heads of business I’ve spoken to in Asia are specifically challenging us to show how generative AI can enhance alpha generation. In other regions, the focus tends to be more on improving productivity, expanding data sources and operational efficiencies.
