AI Research Engineer – Model Compression & Quantization

About the role

Join Tether’s AI research team to lead innovation in model compression and efficient deployment for cutting‑edge multimodal AI systems. You will focus on reducing the footprint and computational cost of large language models (LLMs) and vision‑language models (VLMs) while preserving accuracy, enabling high‑performance AI on resource‑constrained edge devices.

Key responsibilities

• Design and implement compression pipelines using quantization, knowledge distillation and pruning.
• Optimize multimodal architectures that combine text, image and audio inputs.
• Evaluate trade‑offs between model size, latency, and accuracy on edge hardware.
• Collaborate with cross‑functional teams to integrate compressed models into Tether’s products.
• Publish research findings and contribute to open‑source AI tooling.

Required profile

• Strong background in AI research with a focus on model compression techniques.
• Experience working with large‑scale multimodal models such as LLMs and VLMs.
• Proven ability to deliver high‑quality research outcomes in a fast‑paced environment.
• Excellent English communication skills.

Required skills

• Quantization
• Knowledge distillation
• Pruning
• Model compression methods
• Multimodal AI architectures
• Large language models (LLMs)
• Vision‑language models (VLMs)

What we offer

• Fully remote work with a global, high‑impact fintech team.
• Opportunity to shape AI technology for a leading stablecoin ecosystem.
• Access to cutting‑edge research resources and infrastructure.
• Collaborative environment with some of the brightest minds in blockchain and AI.