Multimodal AI Engineer: Vision-Language in Production

Tether.io; United Arab Emirates; On-site; AED 293,793 - 440,690; Full time

Job summary

Tether.io is looking for a passionate member of the AI model team to drive innovation in vision-language models. This role focuses on the full development lifecycle, from data curation to model evaluation and optimization, to create highly capable models for real-world applications.

You will collaborate within a small, talented team, leveraging cutting-edge technology in a setting that values your contributions to push the boundaries of multimodal AI.

Qualifications

• Strong experience with multimodal post-training workflows including supervised fine-tuning.
• Hands-on experience with distributed training frameworks.
• Demonstrated ability to build and improve vision-language models.

Responsibilities

• Conduct end-to-end research and engineering on vision-language models.
• Design and implement evaluation frameworks and benchmarks.
• Build and scale training workflows across distributed GPU infrastructure.

Skills

Multimodal post-training workflows

Parameter-efficient fine-tuning

Reinforcement learning from feedback

Data curation and filtering

Education

Degree in Computer Science, Machine Learning, or a related field

Tools

GitHub

HuggingFace

Job description

Tether.io is looking for a passionate member of the AI model team to drive innovation in vision-language models. This role focuses on the full development lifecycle, from data curation to model evaluation and optimization, to create highly capable models for real-world applications.

You will collaborate within a small, talented team, leveraging cutting-edge technology in a setting that values your contributions to push the boundaries of multimodal AI.