Software Engineer, AI Infrastructure

Location: Bellevue, WA; Palo Alto, CA

Department: Engineering

Software Engineer, AI Infrastructure

About the Role

As a founding AI Infrastructure Engineer, you will report directly to the CTO and lead the development of our core compute and orchestration layer. This is a high-impact role where you will hold a significant ownership stake in the company and lead the 0-to-1 build of our infrastructure. You will work closely with our customers to translate their needs into a world-class platform, while simultaneously shaping our engineering culture and technical direction from the ground up.

What You Will Do

• Architect GPU Compute Fabric: Build and manage the orchestration layer for GPU workloads, ensuring efficient resource allocation and cost management for large-scale training, fine-tuning, and inference.
• Design Developer Interfaces: Build developer-centric SDKs and APIs that transform complex ML workflows into intuitive experiences for researchers and data scientists.
• Operationalize the ML Lifecycle: Develop robust, end-to-end pipelines-from data ingestion and preprocessing to secure model serving and monitoring.
• Client Success & Observability: Work closely with customers to debug fine-tuning jobs and build the observability tools required to track model performance and resource health in real-time.
• Define Systems & Culture Strategy: Lead the technical roadmap by making critical "build vs. buy" decisions on infrastructure and security, while directly shaping the team’s engineering standards and hiring processes.

Qualifications

• 5+ years of software engineering experience, with focus on ML infrastructure or backend systems supporting ML workloads
• Experience deploying and operating ML/DL training or inference pipelines in production (PyTorch, Hugging Face, or similar)
• Hands-on experience with Kubernetes on AWS/GCP, ideally for GPU workloads
• Strong CS fundamentals and system design skills
• Ability to thrive in fast-paced, dynamic environments and navigate ambiguity

About the Company

About Harell Data

Harell Data is building the intelligent infrastructure for the next era of scientific discovery. While AI has revolutionized digital industries, progress in the physical sciences-from drug discovery to material science-is currently bottlenecked because the most valuable training data remains isolated, unorganized, and difficult to access.

We are solving this by providing an end-to-end ecosystem that enables the secure sharing of proprietary datasets alongside the high-performance compute and orchestration layer required to train large-scale models. By bringing compute to the data within a single, integrated environment, we are creating the foundational stack that transforms raw scientific data into the next generation of domain-specific foundation models.