Principal Engineer, Training

CoreWeave is the AI Hyperscaler™, delivering a cloud platform of cutting edge services powering the next wave of AI. Our technology provides enterprises and leading AI labs with the most performant, efficient and resilient solutions for accelerated computing. Since 2017, CoreWeave has operated a growing footprint of data centers covering every region of the US and across Europe. CoreWeave was ranked as one of the TIME100 most influential companies of 2024.

As the leader in the industry, we thrive in an environment where adaptability and resilience are key. Our culture offers career-defining opportunities for those who excel amid change and challenge. If you’re someone who thrives in a dynamic environment, enjoys solving complex problems, and is eager to make a significant impact, CoreWeave is the place for you. Join us, and be part of a team solving some of the most exciting challenges in the industry.  

CoreWeave powers the creation and delivery of the intelligence that drives innovation. 

What You’ll Do:

CoreWeave is seeking a Principal Engineer to be the hands-on technical leader for our next-generation Large-Scale Training Platform. As a senior individual contributor, you will architect and build the fastest, most cost-efficient, and most reliable GPU training services in the industry. You’ll prototype new capabilities, drive engineering standards, and work side-by-side with product, orchestration, and hardware teams to turn CoreWeave’s massive GPU fleet into the best place on earth to train frontier models.

About the role:

• Technical Vision & Strategy - Shape the technical roadmap for distributed training, data-pipeline efficiency, and model-parallel scaling. Evaluate emerging frameworks (DeepSpeed, Megatron-LM, FSDP, Alpa, etc.) and guide build-vs-buy decisions.
• Platform Architecture - Design Kubernetes-native control-plane components that launch and monitor multi-thousand-GPU training jobs. Implement optimizations such as tensor/pipeline parallelism, ZeRO optimizer stages, activation checkpointing, gradient compression, and elastic scaling. Build high-throughput data-ingestion paths (streaming, sharded datasets, RDMA, NVMe-over-Fabric) that keep GPUs saturated.
• Operational Excellence -  Create real-time observability, debugger hooks, and automated rollback / resume for long-running jobs. Develop cost-vs-time-to-train analytics so customers can pick the perfect hardware mix.
• Hands-on Development -  Write production code, reference implementations, and performance benchmarks. Lead design reviews and dive deep into bottlenecks across NCCL, InfiniBand, NVLink, SHARP, and PCIe topologies.
• Mentorship & Collaboration - Coach engineers on large-scale distributed-training best practices. Work directly with lighthouse customers, helping them scale models to tens of billions of parameters.

Who You Are:

• 10+ years building distributed systems or HPC/cloud services, with 4+ years focused on large-scale ML training or similar high-performance workloads.
• Demonstrated expertise in data/model/tensor/pipeline parallelism, mixed-precision training (BF16, FP8), and optimizer sharding (ZeRO, 1-D/2-D ZeRO-Infinity).
• Deep knowledge of PyTorch internals, NCCL/SHARP, RDMA, NUMA, and GPU interconnect topologies.
• Proven track record of squeezing every last TFLOP from multi-node GPU clusters and reducing total training time/cost for massive models.
• Fluency with Kubernetes (or Slurm/Ray) at production scale, plus CI/CD, service meshes, and observability stacks (Prometheus, Grafana, OpenTelemetry).
• Excellent communicator who influences architecture across teams and presents complex trade-offs to executives and customers.
• Bachelor’s or Master’s in CS, EE, or related field (or equivalent practical experience).

Preferred:

• Code contributions to open-source training frameworks (DeepSpeed, Megatron-LM, FSDP, MosaicML, etc.).
• Experience running multi-region or exaFLOP-scale training jobs at a hyperscaler or AI research lab.
• Publications/talks on large-model optimization, gradient compression, or advanced checkpoint orchestration.

Wondering if you’re a good fit? We believe in investing in our people, and value candidates who can bring their own diversified experiences to our teams – even if you aren't a 100% skill or experience match.

Why CoreWeave?

You will be the technical spearhead of an industry-defining training platform, partnering with world-class researchers and engineers who are pushing the boundaries of generative AI and foundation-m