Senior AI Platform Architect — GPU & Inference

Accellor; Mountain View (CA); On-site; USD 180,000 - 280,000; Full time

Job summary

Accellor is seeking a Technical Architect — AI Systems, Inference & Platform Internals to design, scale, and optimize internal AI systems powering ChatGPT and OpenAI API workloads. The role focuses on inference runtime, model serving, GPU infrastructure, and distributed systems for production reliability.

The ideal candidate has 10–12 years of software/ML infra experience, deep knowledge of PyTorch/JAX, and hands-on GPU optimization.

Qualifications

• 10–12 years of software/ML infra experience with strong architecture skills.
• Proven ability to design and operate large AI systems and inference stacks.
• Hands-on expertise across GPUs, model serving, and distributed runtimes.

Responsibilities

• Design and evolve large‑scale AI systems supportingChatGPT/OpenAI API workloads.
• Own latency, throughput, safety, cost, and reliability across platforms.
• Lead inference runtime, model serving, and GPU scheduling at scale.
• Collaborate with research to enable frontier model workloads.

Skills

Python

Distributed systems

GPU CUDA

Model serving

Kubernetes

LLM inference

Context engineering

System architecture

Tools

PyTorch

JAX

TensorFlow

Triton

Job description

Accellor is seeking a Technical Architect — AI Systems, Inference & Platform Internals to design, scale, and optimize internal AI systems powering ChatGPT and OpenAI API workloads. The role focuses on inference runtime, model serving, GPU infrastructure, and distributed systems for production reliability.

The ideal candidate has 10–12 years of software/ML infra experience, deep knowledge of PyTorch/JAX, and hands-on GPU optimization.