AI Systems Architect — Senior Platform Leader

Accellor; San Francisco (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Accellor is seeking a Technical Architect — AI Systems, Inference & Platform Internals to design, scale, and optimize the systems powering AI workloads across ChatGPT, OpenAI API, Codex, and multimodal research tasks.

The role focuses on inference runtime, model serving, GPU infrastructure, context engineering, cost-aware operations, and production reliability, requiring senior hands-on architectural leadership across the full AI platform.

Qualifications

• 10–12 years of software engineering or ML infrastructure experience.
• Hands-on Python and at least one systems/backend language (C++, Go, Rust, Java, or TypeScript).
• Deep understanding of distributed systems, reliability, scalability, and observability.
• Experience with model serving, inference workflows, context engineering, and production deployment.

Responsibilities

• AI Systems Architecture: design large-scale AI systems powering ChatGPT/OpenAI APIs and research workloads.
• Inference Runtime & Model Serving: build high-throughput, low-latency pipelines across GPU clusters.
• GPU, Kernel & Distributed Performance: optimize CUDA/Triton kernels and distributed execution.
• Context Engineering: architect prompt structure, retrieval, and context quality frameworks.
• Cost Optimization: reduce token usage, unnecessary retrieval, and infra spend while preserving quality.
• Release Safety & Validation: define gates, testing, and safe rollout practices.

Skills

Python

C++

Distributed systems

GPU/ CUDA

ML frameworks

Kubernetes

Linux

Tools

CUDA

Docker

Kubernetes

Terraform

Linux

Job description

Accellor is seeking a Technical Architect — AI Systems, Inference & Platform Internals to design, scale, and optimize the systems powering AI workloads across ChatGPT, OpenAI API, Codex, and multimodal research tasks.

The role focuses on inference runtime, model serving, GPU infrastructure, context engineering, cost-aware operations, and production reliability, requiring senior hands-on architectural leadership across the full AI platform.