Senior Compute Network Engineer

Megaport

11d

About The Position

Reporting to the Head of Compute Networking, this role is responsible for the design, build and ongoing evolution of the data centre network fabrics that underpin Latitude's bare metal and accelerated compute platform. You'll work across leaf-spine fabrics, GPU and storage networks, and the automation that keeps them consistent at scale - partnering with internal engineering teams, vendors and integration partners to deliver infrastructure that is fast, predictable and operable. This is the first Compute Networking role based in Australia. You'll be the local anchor point for the region and will work day to day with the existing team in Brazil, so comfort operating across time zones, and a degree of self-sufficiency, matters as much as the technical depth.

Requirements

Passionate about technology and using technology to solve problems
Strong work ethic and willingness to continuously upskill
Strong background in networking and telecommunications
Strong understanding of IP routing, QoS, MTU, resiliency and network failure scenarios.
Practical Experience with BGP, EVPN-VXLAN and leaf-spine architecture.
Experience with large scale DC networks
Experience with large scale MPLS networks
Experience writing scripts to automate tasks
Understanding of high-speed Ethernet, optical transceivers, fibre cabling and data centre cross-connects.
Ability to take initiative and work effectively in a rapidly changing environment
Able to dig into a fault to uncover the root cause

Nice To Haves

Knowledge in programming languages such as Perl and Python
Understanding and exposure to an Agile environment
Understanding of RoCEv2, RDMA, Priority Flow Control and ECN.
Familiarity with Linux networking and infrastructure-as-code practices.
Operational experience working with network datasets, primarily SNMP & sFlow outputs
An interest in data and analysing trends

Responsibilities

Evaluation, testing and certification of new network hardware, systems and software
Design and deploy resilient leaf-spine network fabrics for GPU, AI and high-performance compute infrastructure.
Design and operate IP fabrics using BGP, EVPN-VXLAN and ECMP.
Support GPU frontend, backend, storage, management and out-of-band network connectivity.
Configure and operate switches, routers, and data centre interconnects.
Implement congestion management capabilities, including QoS, ECN and Priority Flow Control where required.
Integrate GPU data centres with Megaport transport, internet transit, cloud and external connectivity services.
Running proof of concept exercises with integration partners
Development of next generation network and systems architectures
Develop network architecture standards, low-level designs, implementation plans and operational documentation.
Maintain technical relationships with vendor engineering teams
Evolving network automation and monitoring systems (Capacity/Latency/Congestion/Availability)
Providing technical support to internal and external customers
Working with internal teams to assist with deployments and operations
Provide technical summaries to enable executive leadership team to make sound business decisions
Last level escalation point for network operations, operations engineering and network engineering teams.