Machine learning · Performance

You make Jane Street's machine learning models train and run faster, tuning everything from GPU code to the network that links the machines.

The seat sits on the firm's growing ML team. Models have to be trained on large amounts of data and then run quickly, both inside live trading systems where delay matters and in research where the aim is to get through as much work as possible. Your job is to find where time is being lost, at any level from a single GPU instruction to the whole cluster, and to remove it.