Senior Systems Engineer, Compute Platform
About Emerald AI
We’re at a pivotal moment for AI and energy. Demand for compute is skyrocketing, but power constraints are becoming a critical bottleneck. Emerald AI sits at the intersection of these two worlds, enabling AI data centers to scale without overwhelming the grid.
Our Emerald Conductor software platform makes data centers flexible and responsive, allowing them to adjust power usage dynamically. This unlocks massive AI growth without major new infrastructure, while also strengthening the grid and supporting the expansion of renewable energy.
We’re a team of experts across AI, cloud, software, and energy - on a mission to scale AI sustainably. We’re backed by leading investors and partners including Radical Ventures and NVIDIA.
Learn more about our vision, team, and backers at https://www.emeraldai.co/.
About the Role
We’re looking for a Senior Systems Engineer, Compute Platform to build and run the compute layer that Emerald Conductor acts on. You’ll bring deep, hands-on experience with virtual machines, networking, storage, and the full instance lifecycle - provisioning, scaling, suspending, migrating, and tearing down workloads across GPU clusters and cloud environments.
Conductor makes AI data centers flexible by changing how and when compute runs. That depends on reliable, well-understood control over the instances underneath: what can be paused or moved, how fast, and with what impact on the people and models relying on them. You’ll design and operate those mechanisms, integrate with the compute platforms our customers and partners use, and make sure every power-driven action is safe, observable, and reversible. You’ll work closely with our software, product, and data center teams, and with infrastructure teams at our customers and partners.
Key Responsibilities
Design, build, and operate the compute platform that Emerald Conductor uses to run and control workloads across GPU clusters and cloud environments
Own the full instance lifecycle - provisioning, scaling, suspending, restore, and teardown—with clear guarantees on speed and workload impact
Build and tune virtualization and orchestration layers (e.g., KVM, Kubernetes, Slurm), including GPU passthrough and partitioning
Design networking for multi-tenant and high-performance workloads, including virtual networks, isolation, and high-bandwidth interconnects such as RDMA
Manage storage for running and paused workloads, including block, file, and object storage, snapshots, and fast checkpointing
Integrate with GPU power and telemetry controls (power capping, clock management, utilization metrics) so Conductor can adjust power safely
Build automation and observability so every power-driven action is monitored, auditable, and reversible
Work with customer and partner infrastructure teams to integrate Conductor with their compute platforms and schedulers
Debug complex issues across hardware, operating systems, networking, and workload layers, and turn findings into durable fixes
Minimum requirements
Bachelor’s degree in computer science, computer engineering, or a related field, or equivalent experience
8+ years building or operating production compute infrastructure, such as cloud platforms, virtualization, or HPC environments
Deep hands-on experience with virtual machines and hypervisors (e.g., KVM/QEMU, libvirt,), including instance lifecycle management at scale
Strong networking fundamentals: TCP/IP, virtual networking, VLANs/VXLAN, load balancing, and network isolation
Strong storage experience across block, file, and object storage, including snapshots and data durability
Strong Linux systems skills and proficiency in at least one programming language (e.g., Go, Python, Rust, or C/C++)
Experience debugging complex issues across hardware, operating system, networking, and application layers
Preferred requirements
Experience at a hyperscaler, GPU cloud provider, or AI lab running large-scale compute
Experience with GPU infrastructure, including NVIDIA drivers, MIG or vGPU, and GPU power and telemetry tools (e.g., DCGM, NVML)
Experience with live migration, checkpoint and restore (e.g., CRIU), or workload preemption
Experience with Kubernetes and HPC schedulers such as Slurm
Familiarity with high-performance networking (InfiniBand, RoCE, RDMA) and distributed storage systems (e.g., Ceph, Lustre, WEKA, VAST, DDN)
Contributions to open-source infrastructure projects
Interest in energy, grid, or data center power systems
What We Offer
Make an impact. Solve the AI power bottleneck and shape how data centers scale sustainably.
Join a world-class team of AI, cloud, software, and energy experts in a collaborative, low-ego environment.
Build from 0→1. Influence strategy, GTM, org design, and customer/investor engagement from day one.
Competitive pay + equity. Stock options let you share in the value you help create.
Comprehensive benefits, including medical, dental, vision, and 401(k) matching.
Flexible location. Work from D.C., Boston, or the Bay Area, with 2 WFH day/week.
Backed by top investors, including Radical Ventures and NVIDIA.
We are an equal opportunity employer and value diversity at our company. We do not discriminate on the basis of race, religion, color, national origin, gender, sexual orientation, age, marital status, veteran status, disability, and any other protected ground of discrimination under applicable human rights legislation.
Emerald AI strives to respect the dignity and independence of people with disabilities and is committed to giving them the same opportunity to succeed as all other employees.
Inclusiveness is core to our culture at Emerald AI, and we strive to ensure you get the most from your interview experience. Emerald AI makes reasonable accommodations for applicants with disabilities. If a reasonable accommodation is needed to participate in the job application or interview process, please reach out to the Talent team.
