Remote AI Kernel & Inference Optimization Engineer

Openkrill; United States; Remote; USD 180,000 - 260,000; Full time

Job summary

Tether is seeking an expert to join its AI model team to advance model serving and inference architectures for multi-modal systems. You’ll optimize deployment strategies for low latency, memory efficiency, and scalability across real-world applications.

You will work on resource-constrained devices and edge platforms, collaborating with researchers and engineers to push the boundaries of practical AI deployment. Candidates should have a PhD in NLP/ML and strong GPU/MSL expertise.

Qualifications

• PhD in NLP or ML with publications in top conferences.
• Expert in mobile GPU compute and MSL (Metal Shading Language).
• Experience optimizing inference latency, throughput, and memory for edge devices.
• Deep understanding of diffusion models and vision transformers.

Responsibilities

• Design and optimize model serving pipelines for high throughput and low latency across devices.
• Develop robust inference tests and benchmarks for edge platforms.
• Collaborate with cross-functional teams to integrate serving frameworks into production.
• Investigate bottlenecks in serving systems and implement improvements.
• Provide empirical research results to push inference performance and efficiency.

Skills

MSL

GPU kernels

Model serving

Distributed inference

Diffusion Models

Vision Transformers

Pruning & Quantization

Education

PhD in NLP

CS degree

Job description

Tether is seeking an expert to join its AI model team to advance model serving and inference architectures for multi-modal systems. You’ll optimize deployment strategies for low latency, memory efficiency, and scalability across real-world applications.

You will work on resource-constrained devices and edge platforms, collaborating with researchers and engineers to push the boundaries of practical AI deployment. Candidates should have a PhD in NLP/ML and strong GPU/MSL expertise.