AI Security Data Scientist — Production ML, Visa Support

Refractal; Greater London; On-site; GBP 90,000 - 130,000; Full time

Benefits offered by this job

Visa sponsorship

Relocation support

Job summary

Refractal, an AI security company, is seeking an experienced ML/Data Scientist to own machine-learning models behind detections in production. You will train, tune and ship classifiers addressing prompt injection, data exfiltration, and policy violations, while designing low-latency models and evaluating performance across customers.

You will analyze production telemetry for drift and false positives, build benchmarks, and collaborate with security engineers to convert new attacks into training

Qualifications

• 5+ years in applied ML or data science with models shipped to production.
• Strong Python, PyTorch or similar, and solid statistics for evaluation.
• Hands-on NLP and LLM work: fine-tuning, embeddings, classification and LLM-as-judge.
• Experience measuring fuzzy problems, including labelling, agreement and drift.
• Clear communication of trade-offs to engineers and customers.

Responsibilities

• Train, tune and ship classifiers for prompt injection, data exfiltration, policy violations and other risky AI activity.
• Design low-latency models, including distilled and small models and rules-plus-ML ensembles.
• Analyse production telemetry for drift, blind spots, false-positive cost and new attack patterns.
• Build evaluations and benchmarks for our detection models, and work with security engineers to turn new attacks into training data.
• Turn company policies and regulatory requirements into measurable controls.

Skills

Applied ML / data science

Python

PyTorch

Statistics

NLP

LLM

Fine-tuning

Embeddings

Classification

Model evaluation

Communication

Equity awareness

Job description

Refractal, an AI security company, is seeking an experienced ML/Data Scientist to own machine-learning models behind detections in production. You will train, tune and ship classifiers addressing prompt injection, data exfiltration, and policy violations, while designing low-latency models and evaluating performance across customers.

You will analyze production telemetry for drift and false positives, build benchmarks, and collaborate with security engineers to convert new attacks into training