Remote AI Research Engineer - Quantization & Compression

Tether.io; Town of Italy (NY); Hybrid; USD 120,000 - 150,000; Full time

Job summary

Tether.io is seeking a talented individual to join their AI model team, focusing on driving innovation in model serving and inference architectures. This role emphasizes optimizing deployment strategies and delivering efficient performance across diverse systems.

Ideal candidates will have a degree in Computer Science (PhD preferred) and extensive experience in AI R&D. Strong knowledge of GPU programming and model optimization is essential, particularly for mobile devices. Join Tether and contribute to revolutionary advancements in digital finance.

Qualifications

• Proven experience in AI R&D with publications in A* conferences.
• Demonstrated ability to apply empirical research to model serving challenges.

Responsibilities

• Design and deploy model serving architectures with high throughput and low latency.
• Monitor key performance indicators in simulated and live environments.
• Prepare high-quality test datasets for real-world deployment challenges.
• Analyze computational efficiency and diagnose serving pipeline bottlenecks.
• Work with cross-functional teams to integrate optimized frameworks.

Skills

Knowledge of Metal Shading Language (MSL)

Low-level kernel optimizations

Inference optimization on mobile devices

Writing GPU kernels for smartphones

Distributed Inference Systems

Mathematics behind Diffusion Models and Vision Transformers

Education

Degree in Computer Science or related field

PhD in NLP, Machine Learning, or related field

Job description

Tether.io is seeking a talented individual to join their AI model team, focusing on driving innovation in model serving and inference architectures. This role emphasizes optimizing deployment strategies and delivering efficient performance across diverse systems.

Ideal candidates will have a degree in Computer Science (PhD preferred) and extensive experience in AI R&D. Strong knowledge of GPU programming and model optimization is essential, particularly for mobile devices. Join Tether and contribute to revolutionary advancements in digital finance.