Senior Network & Site Reliability Engineer

Mxv; San Francisco (CA); On-site; USD 120,000 - 160,000; Full time

Benefits offered by this job

Technical autonomy

Meaningful equity

Elite team of engineers

Job summary

Alembic in San Francisco is seeking a talented Engineer to architect and operate a scalable and secure network infrastructure for machine learning workloads. This role offers the opportunity to build resilient systems and influence architecture decisions in a cutting-edge environment.

Ideal candidates will have extensive experience in network security and a solid background in datacenter operations and network automation. Join us to solve hard problems that impact Fortune 100 companies.

Qualifications

• 8+ years in network or infrastructure engineering, including 5+ years in datacenter operations.
• A strong background in network security, architecture, design, and operations.
• Extensive hands-on experience with network devices and protocols.

Responsibilities

• Architect and operate scalable, secure network architecture.
• Own network device configuration management end to end.
• Improve system and network reliability through automation.

Skills

Network security

Network architecture

Linux systems

Scripting (Python, Bash)

Automation tools (Ansible, Terraform)

Monitoring stacks (Prometheus, Grafana)

Education

8+ years in network or infrastructure engineering

Tools

Network devices (firewalls, switches)

Kubernetes

Monitoring tools (Datadog, ELK)

Job description

About Us

Alembic is the pioneering Causal AI platform. We help the world's largest enterprises move past correlation to prove what actually drives business outcomes — the question marketing and growth teams have never been able to answer with confidence. Fortune 100 companies including Nvidia, Delta Air Lines, and Mars use Alembic to make multimillion-dollar decisions on trusted, causal evidence.

We're backed by a $145M Series B from WndrCo (founded by Jeffrey Katzenberg), Jensen Huang, Joe Montana, Prysm Capital, and Accenture. Our models run on our own NVIDIA DGX SuperPOD built on Grace Blackwell infrastructure — one of the fastest private supercomputers in the world. (We've melted GPUs getting here.)

About the Role

We're building infrastructure that has to perform under real-world scale, reliability, and security demands — and we're looking for an engineer who wants to own the foundation it runs on. This isn't a traditional "keep the lights on" role.

You will design and operate the global network and reliability layer behind one of the world's fastest private supercomputers — the fabric powering distributed compute, ML workloads, real-time analytics, and mission-critical enterprise systems. You'll work across networking, systems, automation, observability, and reliability engineering to scale a platform where performance genuinely matters, with real influence over architecture decisions.

It's a strong fit if you like solving deep infrastructure problems, building resilient systems, automating everything repetitive, and owning architecture rather than just maintaining it.

What You'll Do

• Architect and operate scalable, secure network architecture for high-security requirements and large‑scale machine learning workloads.

• Own network device configuration management end to end, ensuring consistency and reliability across the fleet.

• Improve system and network reliability and performance through automation, observability, and proactive capacity planning.

• Implement and manage complex network protocols and connectivity, including BGP, VPNs, and WAN circuits and external peering.

• Build and maintain comprehensive monitoring, alerting, and incident response — SLOs, runbooks, and on-call rotations — and drive post‑incident analysis and continuous improvement.

• Ensure security, compliance, and operational readiness across our network and cloud infrastructure.

• Partner across engineering and data science to drive a culture of performance and reliability.

What Will Help You Succeed

• 8+ years in network or infrastructure engineering, including 5+ years in datacenter operations and/or systems and network administration.

• A strong background in network security, architecture, design, and operations.

• Extensive hands‑on experience with network devices (firewalls, switches, load balancers) and large-scale architectures and protocols — BGP, QoS, MPLS, and IPsec VPNs.

• Experience designing and operating modern datacenter network fabrics (spine‑leaf, EVPN/VXLAN, ECMP).

• Network automation and IaC tooling (Ansible, Terraform, Nornir, or similar), plus IPAM/DCIM platforms (NetBox, Infoblox, or similar).

• WAN engineering — carrier circuit provisioning and external network peering.

• Familiarity with Kubernetes networking (CNI plugins, ingress, service networking, network policy) and strong operational experience with Linux-based production infrastructure.

• Experience with monitoring and observability stacks (Prometheus, Grafana, Datadog, ELK, OpenTelemetry).

• Solid scripting (Python, Bash) to debug complex network and system issues and automate solutions, plus excellent cross‑functional communication.

Also Helpful

• NVIDIA networking technologies — Cumulus Linux, InfiniBand, Spectrum‑X, and BlueField DPUs (this is the fabric behind our SuperPOD).

• Familiarity with data‑intensive platforms (Spark, Airflow, Kafka) and storage network protocols (NFS, LustreFS, iSCSI).

• Security practices for applications and infrastructure, and experience in high‑compliance or SOC 2 environments.

The Role Is Right for You If

• You want to own mission-critical network and infrastructure end to end — from architecture to incident management — not just keep it running.

• You’d rather build and automate than direct from a distance, and you want meaningful influence over how a high‑performance platform scales.

Why You Might Be Excited About Alembic

• Hard problems with real impact: You'll own the network and reliability layer behind systems that influence multimillion‑dollar decisions at Fortune 100 companies.

• Cutting‑edge technology: Operate our own NVIDIA DGX SuperPOD on Grace Blackwell — one of the fastest private supercomputers in the world — and run a fabric (InfiniBand, Spectrum‑X, BlueField) almost no company has in‑house.

• Technical autonomy: Ownership over architecture decisions and the freedom to solve hard infrastructure problems your way.

• Elite team: Join top engineers who thrive on hard problems and high‑impact work.

• Series B momentum, real ownership: Meaningful equity at a Series B company that's raised $145M, with proven product‑market fit and Fortune 100 traction.

Why You Might Not Be Excited

• If you only want to tell people what to build instead of building and automating alongside them, this isn't the environment for you.

• You prefer companies with 100% built‑out process for every detail.

• You prefer static over dynamic — projects and priorities adapt as we grow. We have real paying customers and a playbook, and we still move at startup speed at Series B scale.