Backbone Network Engineering Principal Job San Francisco area,California USA,IT/Tech

The Experience

Infrastructure Architecture (IA) owns the design, security, and evolution of Salesforce's first-party data center network. We architect the routing, switching, firewall, and load-balancing infrastructure behind large-scale sites that span several Salesforce business units. We also own the backbone connectivity to the internet and to our public cloud environments (AWS and GCP). As a Principal Network Engineer, you will set technical direction for this network and lead the security posture of our edge.

You will also manage the lifecycle of the hardware and designs behind it.

What You'll Actually Be Doing

Define and evolve reference architectures for first-party data center networks. These span routers, switches, firewalls, and load balancers, with lighter touchpoints into compute.
Design network and application delivery architectures for large-scale sites that serve multiple business units across Salesforce.
Own backbone and internet connectivity, plus interconnect between our data centers and public cloud environments (AWS and GCP).
Own infrastructure lifecycle for hardware and designs. For example, you might lead a refresh of an aging top-of-rack switch by evaluating and selecting a suitable newer model, then defining the validated design and transition approach.
Build and maintain network automation and tooling in Ansible and Python, managed in Git repositories. Use AI models in day-to-day design, development, analysis, and documentation work.
Act as a technical authority and escalation point for complex design, capacity, and incident-driven architecture questions.
Lead network security architecture, including security feature design and edge hardening. Partner with Network Operations, Security, Supply Chain/Capacity Planning, and cloud teams to deliver tor engineers and influence cross-org technical strategy through design reviews and architectural documentation.
Drive deployment and tuning of edge protocols and policy, such as mTLS, HTTP/2, and cipher suite changes.
Run software certification and vulnerability and bug management for network platforms, coordinating remediation with vendors and operations teams.

You're Our Person If You Have:

10+ years of experience in network engineering or architecture, with a track record at data center or service-provider scale. Software-driven engineering skills:
Ansible, Python, and Git-based development workflows. Vendor familiarity with Cisco, F5, Juniper, and Ciena.
Familiarity with liquid cooling and high-density facility considerations (such as direct-to-chip cooling, power and thermal constraints, and rack density) and how they influence network hardware selection, rack layout, and design.
AI fluency: hands-on experience and familiarity with AI models, and the ability to apply them effectively in engineering workflows.
Proven record of leading multi-team, multi-year architecture efforts.
Deep expertise in large-scale data center and WAN/backbone design (BGP, EVPN/VXLAN, L2/L3 fabrics, routing policy). Load balancing and application delivery design experience, particularly with F5 (BIG-IP). This includes certificate management and lifecycle, and how certificates and keys are assigned to virtual servers; SSL/TLS termination and re-encryption design, including client-SSL and server-SSL profiles; SSL and HTTP header handling, such as forwarded client IP and protocol headers;

and traffic policy and customization through iRules and related profiles (HTTP, persistence, and others).
Hands-on experience with firewalls and TLS termination at the edge.
Strong security background: vulnerability management, software qualification, and secure protocol configuration. Experience with hardware lifecycle planning, including platform evaluation, refresh, and migration.
Experience with new GPU cluster builds, including topology planning, capacity and scale modeling, and coordination with compute, storage, and facilities teams from design through deployment.
Clear communication with executives and engineers.

Even Better If you have:

Vendor familiarity with Cisco, F5, Juniper, and Ciena.
Familiarity with public cloud connectivity (AWS Direct Connect, GCP Interconnect) and hybrid architectures. Knowledge of AI data center network design, including high-bandwidth, low-latency fabrics for GPU clusters (for example Infini Band or RoCE/Ethernet-based back-end networks) and the front-end/back-end network separation typical of AI workloads.
Familiarity with liquid cooling and high-density facility…