Build the execution plane
for accelerated AI compute.
Vector Fabric is building a provider-independent execution layer for GPU-powered AI workloads—preserving progress, automating failure recovery, and ensuring operational visibility across cloud environments.
Senior ML Infrastructure Engineer
Multiple OpeningsGreater Boston preferred • Remote-friendly • Part-time or Full-time
Part-time structured as an initial 3-month contract; full-time as an ongoing foundational role.
Competitive compensation based on experience and selected engagement structure.
The Role
We are hiring multiple hands-on engineers who can turn emerging product needs into dependable ML infrastructure. We welcome both part-time contractors and full-time candidates, with responsibilities and scope matched to availability and experience.
You will collaborate with our platform engineering team on GPU workload execution, cloud integration, and execution resilience—from design through production validation. This is a systems engineering role, not a model research position.
What You Will Do
- • Own integrations for provisioning, launching, monitoring, and lifecycle management of GPU workloads across supported cloud environments.
- • Translate evolving requirements into practical designs and deliver well-scoped integrations without unnecessary architectural complexity.
- • Run and validate representative ML training, fine-tuning, evaluation, and inference workloads end to end.
- • Develop reliable interfaces for container execution, cloud services, storage, and secure workload data access.
- • Improve execution resilience through job state management, checkpointing, retries, failure handling, and operational visibility.
Qualifications
Required
- • 5+ years of software engineering experience, including 2+ years in ML infrastructure, MLOps, cloud infrastructure, or distributed systems.
- • Strong Python programming, API integration, debugging, testing, and production engineering skills.
- • Hands-on experience running GPU-backed ML workloads using PyTorch, Hugging Face, or similar frameworks; working knowledge of Linux, containers, and object storage.
- • Production experience with at least one major cloud provider (AWS, GCP, or Azure), plus an understanding of workload lifecycles and recovery.
Preferred
- • Experience integrating or extending frameworks such as SkyPilot, Ray, Kubernetes, or Slurm.
- • Experience across specialized GPU providers, cloud portability, workload identity, and secrets isolation.
Ready to apply?
Email us your resume, LinkedIn/GitHub, workload background, and schedule preference (full-time or part-time contract).