XTX Markets is an algorithmic trading firm that actively trades tens of thousands of instruments on over eighty venues with a daily volume of almost three hundred billion USD. Trading algorithms are developed by the quant team, which consists of fewer than twenty researchers and a research cluster of a hundred thousand cores and twelve thousand A/V100 GPUs (and growing fast). The algorithms we develop are necessarily extremely general and robust. The culture is collaborative and friendly. The team structure is flat. There is no bureaucracy.
Trading provides a unique set of challenges for machine learning research. Datasets contain trillions of observations. Markets constantly adapt. They react rapidly, often adversarially, to trades and, over longer horizons, to new trading strategies. Algorithms must be as fast as possible: every microsecond matters when reacting to new market information. At the same time, it is necessary to account for risks associated with holding periods that can be weeks – a range of timescales that spans twelve orders of magnitude.
The models that drive our trading strategies evolved considerably over the last 10 years, from econometric methods that gave a name to the company, to trees, to neural networks, to modern deep learning architectures. We are actively seeking new methods and ideas.