Remote GPU Kernel & Inference Engineer

Anyone Ai Inc.; Neuquén; Remote; ARS 184,482,000 - 292,096,000; Full time

Job summary

Tether is hiring an AI Research Engineer to advance model serving and inference architectures across edge and mobile platforms. You will optimize deployment for low memory footprints and high throughput, tackling latency challenges on resource-constrained devices.

You’ll design robust inference pipelines, measure performance, and drive iterative enhancements in collaboration with cross-functional teams. A PhD or strong research track in ML/NLP is preferred.

Qualifications

• Degree in Computer Science or related field; PhD preferred in NLP/ML or related field.
• Strong background in advanced model architectures and inference algorithms.
• Experience designing and optimizing model serving pipelines for edge and mobile environments.

Responsibilities

• Design and deploy state-of-the-art model serving architectures with high throughput and low latency.
• Build, run, and monitor controlled inference tests in simulated and live environments.
• Identify and prepare test datasets and simulation scenarios for low-resource devices.
• Analyze bottlenecks in serving pipeline and optimize memory and compute usage.
• Collaborate with cross-functional teams to integrate optimized serving frameworks into production pipelines.

Skills

GPU optimization

Model serving

Inference optimization

Low-latency design

Distributed systems

Education

PhD in ML/NLP or related field

BS in Computer Science or related field

Tools

Metal Shading Language (MSL)

Job description

Tether is hiring an AI Research Engineer to advance model serving and inference architectures across edge and mobile platforms. You will optimize deployment for low memory footprints and high throughput, tackling latency challenges on resource-constrained devices.

You’ll design robust inference pipelines, measure performance, and drive iterative enhancements in collaboration with cross-functional teams. A PhD or strong research track in ML/NLP is preferred.