GenAI ML Scientist: Multilingual NLP & Model Training

GCS Recruitment Specialists; Seattle (WA); On-site; USD 140,000 - 210,000; Full time

Job summary

GCS Recruitment Specialists is seeking an experienced Machine Learning Scientist/Engineer to build and fine-tune LLMs, NLP, and multimodal models for production GenAI solutions in a Seattle onsite role. You will work on multilingual content across text, images, and video, train and evaluate models, and collaborate with an established MLOps function to deploy in production.

Candidates should have hands-on experience with fine-tuning, PEFT methods, PyTorch/Hugging Face, and a track record of

Qualifications

• Direct experience fine-tuning transformer models for production use.
• Hands-on training and evaluation of ML/NLP models.
• Experience taking ML models into production environments.
• Strong Python programming skills.

Responsibilities

• Fine-tune and improve foundation models for production GenAI solutions.
• Evaluate model performance and run ablation studies.
• Prepare and curate training datasets for multilingual content.
• Collaborate with MLOps on production deployment.

Skills

Fine-tuning LLMs

Training ML/NLP models

LoRA / QLoRA / PEFT

Hugging Face

PyTorch

Instruction tuning

Dataset curation

Model evaluation & benchmarking

Hyperparameter optimization

NLP / language modeling

Multilingual models

OCR / image understanding

Vision-language / multimodal

Production model deployment

Model hosting/inference

Education

PhD in Computer Science / ML

Master's + 4+ years experience in related field

Tools

PyTorch

Hugging Face

Job description

GCS Recruitment Specialists is seeking an experienced Machine Learning Scientist/Engineer to build and fine-tune LLMs, NLP, and multimodal models for production GenAI solutions in a Seattle onsite role. You will work on multilingual content across text, images, and video, train and evaluate models, and collaborate with an established MLOps function to deploy in production.

Candidates should have hands-on experience with fine-tuning, PEFT methods, PyTorch/Hugging Face, and a track record of