About
Biometric.Vision is an international AI company serving banks, fintech companies, crypto platforms, telecom operators, and government organizations. It develops production-grade AI solutions for Face Biometrics, Digital Identity, KYC, AML, Fraud Prevention, and Identity Verification for global markets.
Responsibilities
•Develop and improve ML/CV models for biometric and identity verification solutions, including Face Verification, Liveness Detection, Anti-Spoofing, and Deepfake Detection
•Prepare, clean, and analyze large datasets; train, test, and enhance ML models
•Integrate modern research approaches and SOTA models into production-grade AI systems
•Optimize model inference and performance across the ML/CV pipeline, from data preparation through deployment
•Collaborate with Backend and other teams during ML solution integration
Requirements
•2+ years of commercial experience in Machine Learning or Computer Vision
•Proficiency in Python and practical experience with PyTorch and/or TensorFlow
•Good understanding of Computer Vision and Deep Learning, including work with images and large datasets
•Understanding of ML model training, validation, and evaluation processes, plus experience with model optimization or inference
Nice to have
•Experience with Face Biometrics, KYC, Identity Verification, Anti-Fraud, Liveness Detection, Anti-Spoofing, Face Recognition, or Deepfake Detection
•Experience using OpenCV, InsightFace, DeepFace, ONNX Runtime, TensorRT, or CUDA
•Docker/Kubernetes
•AWS/GCP
•MLOps/CI/CD
•Experience implementing research papers and SOTA models in production or participating in biometric certification and evaluation projects
Conditions
•Work location: Almaty
