AI Engineer/ML Engineer - Senior Developers - AI Training - San Diego, USA

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

Benefits offered by this job

Fully remote work

Flexible hours

Job summary

Prolific Academic Ltd is offering a remote, research-oriented role for experienced AI/ML engineers to train and evaluate language models. Contributors work remotely with flexible blocks for technical review and evaluation tasks, typically in one-hour focus sessions.

The position suits senior engineers in the United States who can commit to focused blocks of ML review and provide high-quality feedback for RLHF alignment.

Qualifications

• BS, MS, or PhD in Computer Science, AI, Robotics, or related field with ML focus.
• Experience building, deploying, or fine-tuning ML models in production.
• Deep familiarity with Transformers, CNNs, and RNNs; optimization techniques.
• Hands-on with prompt engineering, RLHF, or retrieval-augmented generation (RAG).
• Attention to detail for spotting hallucinations and biased outputs.

Responsibilities

• Review explanations of model architectures, loss functions, and backpropagation for accuracy.
• Audit ML code and notebooks: training loops, preprocessing, evaluation.
• Provide high-quality human feedback for RLHF alignment.
• Analyze model reasoning on chain-of-thought prompts and flag issues.
• Conduct comparative testing across outputs using defined criteria.

Skills

ML model training

Prompt engineering

RLHF pipelines

RAG systems

Education

BS/MS/PhD in CS/AI/Robotics

Tools

PyTorch

TensorFlow/Keras

Job description

Role overview

This participant-track opportunity invites experienced AI and machine learning engineers to help train and evaluate large language models through structured paid research tasks. Instead of a conventional employment arrangement, contributors work remotely on a flexible basis, lending their professional judgment to sharpen model behavior. The position is well suited to senior engineers in the San Diego area who can commit to focused blocks of technical review.

Responsibilities

• Review AI-generated explanations of model architectures, loss functions, and backpropagation for technical accuracy.
• Audit machine learning code and notebooks, covering training loops, data preprocessing scripts, and evaluation routines.
• Supply high-quality human feedback used in reinforcement learning from human feedback (RLHF) to align models with human intent and safety.
• Analyze how models navigate complex chain-of-thought prompts and flag breakdowns in their reasoning.
• Conduct comparative testing across model outputs using defined technical taxonomies and performance criteria.

Requirements

• BS, MS, or PhD in Computer Science, Artificial Intelligence, Robotics, or a related quantitative field with a machine learning emphasis.
• Professional experience building, deploying, or fine-tuning ML models in a production environment.
• Deep familiarity with neural network architectures such as Transformers, CNNs, and RNNs, including optimization techniques.
• Hands-on work with prompt engineering, RLHF pipelines, or retrieval-augmented generation (RAG) systems.
• Sharp attention to detail for catching hallucinations, biased outputs, or logical failures in AI-generated technical content.
• Capability to audit complex model logic, evaluate mathematical proofs behind ML algorithms, and detect potential training data contamination.

Nice to have

• Expert proficiency in PyTorch or TensorFlow/Keras.
• Advanced Python including NumPy, Pandas, Scikit-learn, and Hugging Face Transformers.
• Experience with AWS SageMaker, Google Vertex AI, or tools such as Weights & Biases and LangChain.
• Familiarity with vector databases like Pinecone, Milvus, or Weaviate for RAG evaluation.

Benefits and work setup

• Hourly compensation up to $80 for qualifying contributors.
• Fully remote work with flexible hours and the option to work from home.
• Brief onboarding through a 10 to 15 minute skills assessment.
• Tasks typically require one hour of uninterrupted focus, though many are shorter.