AI Engineer/ML Engineer - Senior Developers - AI Training - Brighton, UK

prolificacademicltd; United States; Remote; USD 66,000 - 110,000; Part time

Job summary

Prolific Academic Ltd is seeking an expert in machine learning to participate in paid contract work evaluating and training large language models. You will review model architectures and training components, provide structured feedback, and help ensure high-quality reasoning and safety across tasks.

The role offers fully remote scheduling with study-based tasks, typically up to one hour each, and compensation up to $80 per hour depending on study requirements.

Qualifications

• Bachelor's, Master's or PhD in Computer Science, AI, Robotics or related quantitative field with ML focus.
• Hands-on production experience building, deploying, or fine-tuning ML models.
• Strong understanding of neural networks incl. Transformers, CNNs and RNNs with optimization knowledge.
• Experience with prompt engineering, RLHF or retrieval-augmented generation (RAG).
• Ability to audit model logic and assess training-data quality for correctness.

Responsibilities

• Review explanations of architectures, loss functions, and backpropagation for technical accuracy.
• Audit ML code, training loops, preprocessing scripts, and evaluation notebooks for efficiency and accuracy.
• Provide structured feedback for RLHF workflows and model alignment tasks.
• Assess how models handle complex chain-of-thought prompts and identify reasoning gaps.
• Benchmark model outputs against defined technical taxonomies and metrics.

Skills

ML model development

Attention to detail

RLHF workflows

Model evaluation

Prompting

Education

BS/MS/PhD in CS/AI/ML

Tools

PyTorch

TensorFlow/Keras

NumPy/Pandas/Scikit-learn

Hugging Face Transformers

Cloud ML platforms

Job description

Role overview

Join an expert network of machine learning professionals contributing to the training and evaluation of large language models. This is a participant-based, paid contract role based in Brighton where your technical expertise directly shapes AI model accuracy, safety, and reasoning quality. Successful candidates complete paid study tasks, typically requiring up to one hour of focused work, on a flexible remote schedule.

Responsibilities

• Review AI-generated explanations of model architectures, loss functions, and backpropagation for technical correctness
• Audit machine learning code, training loops, preprocessing scripts, and evaluation notebooks for efficiency and accuracy
• Provide structured human feedback that supports reinforcement learning from human feedback (RLHF) workflows and model alignment
• Critically assess how models navigate complex chain-of-thought prompts and pinpoint where reasoning breaks down
• Benchmark and compare model outputs against defined technical taxonomies and performance metrics

Requirements

• BS, MS, or PhD in Computer Science, Artificial Intelligence, Robotics, or a related quantitative field with a machine learning focus
• Hands-on production experience building, deploying, or fine-tuning ML models
• Professional-level understanding of neural network architectures such as Transformers, CNNs, and RNNs, plus optimization techniques
• Practical experience with prompt engineering, RLHF, or retrieval-augmented generation (RAG) workflows
• Ability to audit model logic, flag potential training-data contamination, and evaluate the mathematical proofs behind ML algorithms
• Strong attention to detail for spotting hallucinations, biased outputs, or logical failures in AI-generated technical content

Nice to have

• Proficiency with PyTorch or TensorFlow/Keras
• Advanced Python skills including NumPy, Pandas, Scikit-learn, and Hugging Face Transformers
• Experience with cloud MLOps platforms such as AWS SageMaker, Google Cloud Vertex AI, Weights & Biases, or LangChain
• Familiarity with vector databases like Pinecone, Milvus, or Weaviate for RAG evaluation

Benefits and work setup

• Hourly compensation commonly up to $80, depending on study requirements
• Fully remote with flexible scheduling aligned to available studies
• Short 10-15 minute skills assessment as the entry point, with quick onboarding once accepted