Systems Engineer

We are in need of a smart, creative yet careful engineer to take over primary responsibility for our complex network of servers and systems that support core research and trading.  This engineer will be a collaborator in designing and implementing end-to-end solutions across groups.

Responsibilities include:

  • Design, build, and maintain the growing Linux compute cluster that is the backbone of our trading research efforts
  • Design, build, and maintain ultra-low-latency systems that mirror our production trading machines
  • Manage and tune the cluster network (Ethernet and Infiniband)
  • Maintain authentication and access-control systems (Kerberos, NIS, etc)
  • Perform systems tuning, including both hardware and low-level kernel tuning
  • Manage the hardware lifecycle and procure new hardware as necessary
  • Automate common tasks, including overnight batch processing jobs
  • Evaluate / experiment with new, cutting-edge hardware

Skills / Experience:

  • Expert-level knowledge of Linux environments
  • Experience in an HPC / distributed-computing environment
  • Experience as an administrator within a high-value production environment
  • Familiarity with Rocks / StackIQ or similar
  • In-depth knowledge of TCP/IP, including multicast, Ethernet, and Infiniband
  • In-depth knowledge of the Linux kernel, including its IRQ and scheduling algorithms
  • Ability to tune a system for ultra-low-latency
  • Experience with Mellanox hardware
  • Familiarity with kernel-bypass techniques and hardware (e.g., Solarflare)
  • Familiarity with pros and cons of the various filesystems (XFS, ZFS)
  • Ability to manage a complex storage hierarchy (including various RAID arrays)
  • Ability to automate common tasks with Python or Bash scripting
  • Excellence in communication and “owning” a task from conception to completion