Software Engineer - Host Networking

Socket.dev; Bellevue (CA); On-site; USD 184,000 - 257,000; Full time

Job summary

Meta is seeking software engineers to design, develop, and validate drivers, firmware, and software for network devices and AI workloads. You will debug system-level issues, optimize performance, and lead integration across partner teams.

You will also develop automated test suites for CI/CD, collaborate to integrate components, and contribute to on-call rotations. The role covers datacenter switches, HPC libraries, and scalable communication infrastructure.

Qualifications

• Bachelor's degree in Computer Science, Computer Engineering, or related field, or equivalent practical experience.
• 7+ years of software development experience, or 4+ years with a PhD.
• 3+ years of experience with network devices (routers, switches, adapters, load balancers) and understanding of routing protocols.
• Proficiency in C/C++ and Python or Shell scripting.
• Experience developing and automating test suites.
• Experience collaborating with cross-functional teams (hardware, firmware, test) to deliver technical solutions.

Responsibilities

• Design, develop, and validate drivers, firmware, and software for network devices, transport stacks, and AI workloads.
• Debug complex system-level issues and optimize software stack performance.
• Understand software components from multiple partner teams, lead integration efforts, and drive continued development.
• Develop and automate test suites for CI/CD framework and various components.
• Collaborate with partner teams to integrate software components, align on goals, and participate in oncall rotations.
• Design, develop, and deploy services to manage datacenter network switches and forwarding functions.
• Enhance HPC libraries such as NCCL, RCCL, OneCCL, MPI.
• Develop infrastructure to maintain switch state and enable communication between services at scale.

Skills

C/C++

Python

Test automation

CI/CD

Education

Bachelor's degree in CS

Job description

At Meta, we're building and operating one of the world's most dynamic and fast-paced networks, powering our global data centers and supporting advanced technologies like AI, Generative AI, Recommendation engines, and Metaverse. Our network infrastructure teams are responsible for developing, deploying, and operating this complex system, covering the entire network lifecycle from hardware development to operation. We're seeking software engineers to join our teams and help build scalable distributed systems, develop effective solutions to our challenges, and ship them into production. As part of our network engineering teams, you'll have the opportunity to work on advanced switching technology, collaborate with engineers, and contribute to the development of Meta's hyper-scale network infrastructure.

Responsibilities

• Design, develop, and validate drivers, firmware, and software for network devices, transport stacks, and AI workloads
• Debug complex system-level issues and lead performance tuning exercises to optimize software stack performance
• Understand software components from multiple partner teams, lead integration efforts, and drive continued development
• Develop and automate test suites for CI/CD framework and various components
• Collaborate with partner teams to integrate software components, align on goals, and participate in oncall rotations
• Design, develop, and deploy services to manage datacenter network switches and forwarding functions
• Enhance HPC collective communication and parallel computing libraries (NCCL, RCCL, OneCCL, MPI)
• Develop infrastructure to maintain switch state and enable communication between services at scale

Minimum Qualifications

• Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience
• 7+ years software development experience in industry settings or PhD with 4+ years of experience
• 3+ years relevant experience with network devices and products (routers, switches, adapters, load balancers) and an understanding of network routing protocols in industry settings
• Proficiency in C/C++ and at least one scripting language (Python/Shell Scripting)
• Experience with developing and automating test suites
• Experience collaborating with cross-functional teams (e.g., hardware, firmware, test) to define and deliver technical solutions

Preferred Qualifications

• Experience with one of: platform services (program, control, and monitor Optics, Physical Layer (PHY), FPGAs, sensors, fan control, power etc), Board Support Package (BSP), Operating Systems, Kernel, Bootloader, Power Management, Real-Time Operating System (RTOS), Linux
• Working knowledge of transport stack particularly Remote Direct Memory Access (RDMA) and/or RDMA over Converged Ethernet version 2 (RoCEv2)
• Parallel computing platforms such as CUDA, RoCM and OpenCL
• Linux Kernel, especially drivers and network stack
• Qemu, FPGA Emulation environment is a plus
• Systems programming, TCP/IP, HTTP/HTTPS, SPDY, DNS, and load balancers

$183,997/year to $257,000/year + bonus + equity + benefits