AI Infrastructure Architect

Huayun North America, Inc. (Maxta); Palo Alto (CA); Hybrid; USD 210,000 - 300,000; Full time

Benefits offered by this job

Equity in pre-IPO company

Hybrid work model (Sylcon Valley + REM

Health, dental, vision + 401K matching

Conference budget + continuouslearning

Patent portfolio exposure

Direct work with CEO

Job summary

Maxta is seeking an AI Infrastructure Architect to design the core MaxtaOS and MaxModel platforms in Palo Alto. You will drive the technical vision for heterogeneous GPU abstraction, intelligent compute scheduling, and model deployment pipelines for enterprise-scale workloads.

Lead the integration with edge devices, define API boundaries, and mentor the team while collaborating with leadership. This role offers equity, hybrid work in Silicon Valley, and exposure to cutting-edge AI systems.

Qualifications

• 8+ years in distributed systems, infrastructure, or platform engineering.
• Deep expertise in GPU computing (CUDA, multi-GPU scheduling, heterogeneous hardware).
• Production experience with Kubernetes or custom scheduling systems.
• Strong background in systems programming languages (Python, Go, or C++).

Responsibilities

• Architect MaxtaOS core systems: GPU resource management and multi-tenant scheduling.
• Design MaxModel registry and deployment pipeline: packaging, version control, AB testing infra.
• Lead integration with edge computing platforms (Jetson, custom CUDA kernels).
• Define API contracts and system boundaries across MaxtaOS+MaxModel stack.
• Mentor engineering team on distributed systems best practices.

Skills

Distributed systems

GPU computing

Python/Go/C++

Model serving

Edge computing

System design

Technical documentation

Tools

Kubernetes

TorchServe

Triton

Jetson

CUDA

Job description

Description

As Maxta's AI Infrastructure Architect, you will design and build the core architecture of MaxtaOS and MaxModel — our enterprise AI operating system and model governance platform. You will lead the technical vision for heterogeneous GPU abstraction, intelligent compute scheduling, and model deployment pipelines that serve Fortune 500 enterprises.

What You Will Do

• Architect MaxtaOS core systems: GPU resource management, multi-tenant scheduling, hardware abstraction layer
• Design MaxModel registry and deployment pipeline: model packaging, version control, A/B testing infrastructure
• Lead integration with edge computing platforms (Jetson, custom CUDA kernels)
• Define API contracts and system boundaries across the MaxtaOS + MaxModel stack
• Mentor engineering team on distributed systems best practices

Requirements

• 8+ years in distributed systems, infrastructure, or platform engineering
• Deep expertise in GPU computing (CUDA, multi-GPU scheduling, heterogeneous hardware)
• Production experience with Kubernetes, container orchestration, or custom scheduling systems
• Strong background in Python, Go, or C++ for systems programming
• Experience with ML model serving (TorchServe, Triton, vLLM, or similar)
• Familiarity with edge computing platforms (NVIDIA Jetson, ARM64 deployment)
• Excellent system design and technical documentation skills

Benefits

• Competitive salary + equity in a pre-IPO AI infrastructure company
• Work directly with the CEO on cutting-edge AI systems
• Backed by Andreessen Horowitz and Intel Capital
• 100+ patents portfolio — real deep-tech, not wrapper products
• Flexible hybrid work model (Silicon Valley + remote)
• Health, dental, vision + 401K matching
• Conference budget + continuous learning