Top Tech Jobs & Startup Jobs

The Senior Treasury Analyst will manage banking relationships, optimize cash flow, ensure compliance with financial regulations, and prepare treasury reports.

The Senior Power Consultant will manage power scheduling, demand response strategies, power procurement, contract negotiations, regulatory compliance, and market intelligence to optimize energy costs for Bitcoin mining.

Lead the architecture and implementation of AI workload scheduling and orchestration for a large-scale NeoCloud platform. Build batch and gang scheduling, admission control, queueing, topology-aware placement, GPU sharing, and multi-tenancy capabilities using Kubernetes technologies. Optimize NVLink and InfiniBand utilization, integrate scheduling with hardware and storage systems, improve reliability in HPC environments, resolve contention and deadlocks, and mentor engineers while guiding architectural decisions.

Architects and operates high-performance storage infrastructure for AI training and inference. Responsibilities include developing Kubernetes CSI drivers, integrating GPUDirect Storage, optimizing NVMe caching and I/O performance, supporting RDMA and high-speed networking, monitoring storage systems, managing quotas and multi-tenancy, and mentoring engineers.

Architects and operates the hardware foundation for AI cloud infrastructure, integrating NVIDIA and AMD GPUs with Kubernetes. Configures high-performance networking using RDMA, SR-IOV, RoCEv2, and InfiniBand; automates GPU and NIC remediation; manages GPU slicing, kernel and driver tuning, bare-metal provisioning, and firmware updates. Investigates cross-layer performance issues, supports topology-aware scheduling and storage, establishes operational standards, and mentors engineers.

Designs and integrates distributed and parallel storage systems into a multi-region GPU cloud. Responsibilities include storage provisioning, mounting, tenant isolation, quotas, lifecycle management, I/O performance tuning, replication, consistency, durability, monitoring, capacity planning, and operational runbooks. The engineer also owns golden images, GPU drivers, CUDA, templates, container registries, and multi-region release distribution while collaborating with compute, networking, and control-plane teams.