Remote AI/ML Engineer — Train & Fine‑Tune LLMs

Prolific Academic Ltd; United States; Remote; USD 8,266,000 - 11,021,000; Full time

Benefits offered by this job

Competitive pay

Remote work from home

Job summary

Prolific is seeking AI/ML Engineers to join our Expert Network, helping train and evaluate next-generation LLMs. You’ll tackle high-skill tasks and complete a brief test to assess fit, with paid opportunities for focused one-hour tasks.

This role offers remote work from anywhere in the US, with competitive pay up to $80/hour and flexible hours for experienced professionals who can audit model logic and drive improvements in RLHF and RAG workflows.

Qualifications

• Advanced degree in CS/AI/Robotics with ML focus.
• Experience building, deploying, or fine-tuning ML models in production.
• Deep learning mastery: transformers, CNNs, RNNs, optimization.
• LLM workflows: RLHF or RAG.
• Ability to audit complex model logic and data quality.
• Attention to detail in spotting hallucinations and biases.

Responsibilities

• Evaluate LLM architecture logic with technical rigor.
• Audit ML code, training loops, and evaluation notebooks.
• Refine RLHF frameworks to align models with human intent and safety.
• Analyze model reasoning and identify failure points in chain-of-thought prompts.
• Benchmark performance across model outputs using defined metrics.

Skills

Deep learning mastery

LLM specialization

Analytical critique

Code and notebook audit

Model evaluation & benchmarking

Education

BS/MS/PhD in CS/AI/Robotics with ML focus

Tools

PyTorch

TensorFlow/Keras

Hugging Face Transformers

NumPy & Pandas

Job description

Prolific is seeking AI/ML Engineers to join our Expert Network, helping train and evaluate next-generation LLMs. You’ll tackle high-skill tasks and complete a brief test to assess fit, with paid opportunities for focused one-hour tasks.

This role offers remote work from anywhere in the US, with competitive pay up to $80/hour and flexible hours for experienced professionals who can audit model logic and drive improvements in RLHF and RAG workflows.