Remote AI Research Engineer: Model Compression & Multimodal

Tether.io; Ireland; On-site; EUR 90,000 - 150,000; Full time

Job summary

Tether.io is seeking an AI research engineer to advance compression techniques for multimodal models, including LLMs and VLMs. You will focus on reducing model footprint and deployment costs while preserving accuracy, enabling efficient edge deployment and scalable AI services.

The role emphasizes quantization, distillation, and pruning, with opportunities to publish findings and contribute to cutting-edge fintech AI research.

Qualifications

• PhD in NLP/ML or related field with strong publications.
• Experience with PyTorch and deep learning.
• Hands-on with model compression (quantization, distillation, pruning).

Responsibilities

• Apply low-bit quantization to reduce model size and latency for multimodal models while maintaining accuracy.
• Leverage knowledge distillation to transfer capabilities from larger models to smaller ones.
• Implement pruning techniques to remove redundant parameters and attention heads.
• Analyze trade-offs between model efficiency and accuracy; propose improvements.
• Document methodologies and results to support reproducibility.

Skills

English communication

Multimodal AI research

Edge deployment

Quantization-aware training

Post-training quantization

Education

PhD in NLP/ML or related field

Bachelor's in Computer Science or related field

Tools

PyTorch

C++

Knowledge distillation

Model pruning

Quantization

Transformers

Job description

Tether.io is seeking an AI research engineer to advance compression techniques for multimodal models, including LLMs and VLMs. You will focus on reducing model footprint and deployment costs while preserving accuracy, enabling efficient edge deployment and scalable AI services.

The role emphasizes quantization, distillation, and pruning, with opportunities to publish findings and contribute to cutting-edge fintech AI research.