Applied Scientist II - Computer Vision for Identity Verification

Entrust; Greater London; Hybrid; GBP 90,000 - 130,000; Full time

Benefits offered by this job

25 days annual leave

Birthday day off

Pension

Private medical & dental insurance

Life enrichment allowance

Job summary

Entrust in London is seeking an Applied Scientist II to design and train cutting-edge ML solutions for digital identities. You will work on deepfake detection, bias mitigation, document understanding, and efficient ML while collaborating with product and engineering to deploy state-of-the-art identity-focused products.

The role requires strong Python and PyTorch skills, a passion for fair ML, and the ability to publish research results at top conferences.

Qualifications

• Strong experience in machine learning and computer vision.
• Proven track record delivering high-performance ML-driven products.
• Deep understanding of ML theory.
• Strong coding skills in Python and PyTorch.
• Care about building fair and cutting-edge ML products.

Responsibilities

• Push the frontier of research in deepfake detection, bias mitigation, fraud/anomaly detection, face matching, document understanding, and efficient on-device ML.
• Publish research results in national and international conferences and journals; collaborate with product and engineering to improve identity-focused products.
• Implement bias-mitigation strategies to build fair models.
• Train and benchmark large-scale vision-language models for document extraction.
• Train a multi-modal document understanding model from scratch using synthetic data.
• Optimize LoRA adapter latency in PEFT/Triton.
• Profile, debug and improve model training speed on multiple GPUs.
• Experiment with multimodal models to detect fraud.

Skills

Machine learning

Computer vision

Python

PyTorch

Tools

AWS

Ray

Weigths & Biases

Job description

Entrust in London is seeking an Applied Scientist II to design and train cutting-edge ML solutions for digital identities. You will work on deepfake detection, bias mitigation, document understanding, and efficient ML while collaborating with product and engineering to deploy state-of-the-art identity-focused products.

The role requires strong Python and PyTorch skills, a passion for fair ML, and the ability to publish research results at top conferences.