Network Engineer, AI/ML Infrastructure na Bosonai

About The RoleWe're seeking an experienced Network Engineer to design, build, and optimize the high-performance networking infrastructure powering our AI/ML operations in Toronto. You'll work at the cutting edge of network technology—managing InfiniBand and ultra-high-speed Ethernet fabrics that connect NVIDIA H100 and A100 GPUs, over 20PB of Ceph storage, and hundreds of servers.You'll be hands-on with the full lifecycle of our network infrastructure: planning, building, testing, deploying, and keeping everything running at peak performance. That means troubleshooting issues as they arise, monitoring network performance and throughput, developing automation to streamline operations, and working closely with HPC and ML teams to ensure they have the bandwidth they need. You'll also help us plan for future capacity and evaluate emerging network technologies as we scale to meet increasingly demanding workloads.

Responsibilities

Configure and maintain InfiniBand and high-speed Ethernet fabrics

Optimize network performance for RDMA, and GPU-to-GPU communication

Manage network switches (Mellanox, NVIDIA, Micas Networks)

Troubleshoot network bottlenecks and latency issues

Plan and execute network upgrades and expansions

Network security implementation (firewalls, VLANs, ACLs)

Collaborate on storage network optimizationInfrastructure monitoring

Minimum Qualifications

4+ years of network engineering experience in production environments

Strong understanding of L2/L3 networking protocols (TCP/IP, BGP, OSPF, VLANs)

Hands-on experience with high-speed networking (100Gb+ Ethernet and InfiniBand)

Hands-on experience with network security (firewalls, ACLs, network segmentation)

Knowledge of HPC network topologies

Experience with InfiniBand fabrics including RDMA, RoCE, IPoIB

Strong troubleshooting and problem-solving skills

Preferred Qualifications

Experience in data center environments or AI/ML infrastructure

Hands-on experience with high-performance Ethernet switches (e.g., Broadcom Tomahawk), and latest InfiniBand switches (e.g., Nvidia/Mellanox)

Experience optimizing networks for GPU-to-GPU communication

Experience with open-source firewall solutions (OPNsense, pfSense, or similar)

Experience with network automation tools

Understanding of distributed storage networking (Ceph cluster networks)

Familiarity with network monitoring and observability tools (Prometheus, Grafana)

Knowledge of multi-site network connectivity and WAN optimization

Familiarity with cloud networking in at least one platform (AWS, GCP, or Azure) including VPC design, site-to-site VPN configuration, Direct Connect/ExpressRoute/Cloud Interconnect, hybrid cloud connectivity, and cloud-to-datacenter network integration

Candidatar-se agora