Senior ML Engineer: Real-Time Verification & Anomalies

Sumsub; España; Presencial; EUR 75.000 - 105.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Remote-first culture

Flexible work hours

Extra time off

Descripción de la vacante

Sumsub, a leading full-cycle verification platform, is seeking an ML/CV Engineer to own end-to-end model development in a high-load production environment. You will train, deploy, and monitor models that detect deepfakes, segment documents, and identify anomalies in real time, using PyTorch and FastAPI within a robust stack.

You will collaborate on data labeling, experiments, and scalable pipelines, working with PMs and backend teams worldwide.

Formación

• 4+ years of experience as an ML/CV engineer in fast-paced, high-load product environments.
• Strong background in deep learning frameworks (PyTorch) and production-grade engineering skills.
• Fluent with data side work: labeling, collection, and navigating data space for training.
• Familiar with current research, baselines, and modern architectures.
• Autonomous product owner capable of taking ideas from inception to implementation.
• Proactive communicator who aligns with PMs and backend teams.

Responsabilidades

• Own the full ML lifecycle: data collection, labeling, training through optimization, production rollout, and ongoing quality assessment.
• Train CV/ML models to detect deepfakes, segment documents, and uncover anomaly patterns in real time.
• Balance research and speed by choosing full models or clever heuristics for fast solutions.
• Serve models with FastAPI and PyTorch, integrating with a high-load Java backend.
• Build and monitor ML pipelines and dashboards with Airflow, ClickHouse, and Superset to maintain reliability under load.

Conocimientos

PyTorch

Deep learning

MLOps

Data labeling

Communication

Herramientas

FastAPI

Airflow

ClickHouse

Superset

Triton

Descripción del empleo

Sumsub, a leading full-cycle verification platform, is seeking an ML/CV Engineer to own end-to-end model development in a high-load production environment. You will train, deploy, and monitor models that detect deepfakes, segment documents, and identify anomalies in real time, using PyTorch and FastAPI within a robust stack.

You will collaborate on data labeling, experiments, and scalable pipelines, working with PMs and backend teams worldwide.