Machine Learning & NLP Expert

Role overview

Independent contractor role for an ML & NLP specialist supporting an AI research initiative. You will design real-world ML/NLP tasks, produce high-quality reference solutions, and evaluate frontier AI model performance to improve reasoning and overall capability.

Responsibilities

• Design challenging machine learning and natural language processing tasks, including:
  • Model development and evaluation
  • Natural Language Understanding (NLU)
  • Natural Language Generation (NLG)
  • Information Retrieval and Search
  • Applied machine learning pipelines
  • Transformer models and Large Language Models (LLMs)
• Develop accurate reference solutions and integrate tasks into agentic development environments using Python.
• Build executable evaluation frameworks and testing components where appropriate.
• Evaluate model outputs for:
  • Technical correctness
  • Reasoning quality
  • Overall performance
• Identify capability gaps and classify model failure modes; provide detailed written analyses.
• Create and refine evaluation guidelines, scoring rubrics, and quality standards.
• Collaborate with other subject-matter experts to ensure consistency and high-quality training data.

Requirements

• Deep hands-on experience in ML and/or NLP via industry, research, or graduate/PhD-level work.
• Strong Python proficiency for developing ML/NLP applications.
• Strong understanding of modern ML techniques, including:
  • Model training and evaluation
  • Transformer architectures
  • Large Language Models (LLMs)
  • NLP pipelines
  • Feature engineering and model optimization
• Experience with one or more of:
  • PyTorch
  • TensorFlow
  • Hugging Face Transformers
  • Equivalent frameworks
• Ability to commit ~20 hours/week.
• Excellent written communication and ability to work independently in a remote setting.

Preferred qualifications

• Experience with AI model evaluation, AI training data creation, or human-in-the-loop assessment.
• Familiarity with RAG (Retrieval-Augmented Generation), vector databases, embedding models, or multimodal AI.
• Experience building benchmarking frameworks and automated evaluation/testing pipelines.
• Open-source ML/NLP contributions or published research.
• Experience with production-scale ML systems.

Logistics

• Work arrangement: fully remote
• Schedule: ~20 hours per week
• Duration: project-based, may extend based on performance and requirements
• Compensation: $80–$110 per hour (paid weekly via Stripe or Wise)