Machine Learning Engineer

Qantev; Paris; Sur place; EUR 60 000 - 90 000; Plein temps

Résumé du poste

A leading AI platform firm in Paris seeks a Machine Learning Engineer to develop next-generation AI models. The role involves creating deep learning architectures focusing on NLP and fraud detection, contributing to innovative healthcare solutions. Ideal candidates will have strong expertise in PyTorch or TensorFlow, and experience in MLOps practices, enhancing product performance through collaboration with various teams.

Qualifications

• 5+ years in machine learning or deep learning engineering.
• Expert in PyTorch or TensorFlow.
• Strong background in NLP and vision-language models.

Responsabilités

• Architect and develop deep learning models for NLP and anomaly detection.
• Build domain-specific LLMs for document OCR and medical inference.
• Design model-serving pipelines with monitoring.

Connaissances

Machine Learning

Deep Learning

NLP

Model Optimization

MLOps

Communication

Outils

PyTorch

TensorFlow

Hugging Face

Description du poste

Qantev is the most advanced AI Platform dedicated to helping health insurers deliver superior healthcare and claims experience to their members.

By leveraging insurers' historical health claims data and applying advanced Machine Learning techniques and Generative AI, Qantev predicts patient journeys, optimizes healthcare outcomes and streamlines healthcare payers operations.

Founded in 2018 and backed by top investors and industry leaders, we are a team of over 80 talented and diverse professionals based in Paris and Hong Kong, serving clients across Europe, the United States, Latin America, Asia, and the Middle East.

Together, we are revolutionizing the insurance claims landscape with AI-driven solutions

About the Role

You will shape Qantev's next-generation AI models, from information extraction to anomaly detection.

Your work will power document understanding, medical code inference, and scalable fraud detection architectures.

You will occupy a central position in the model development lifecycle, writing high-quality production code for our inference server, aligning closely with the product and platform teams,

and designing automated evaluation pipelines to continuously assess and improve performance.

Responsibilities

• Model Design : Architect and develop deep learning models for NLP (transformers, VLMs), anomaly detection (GNNs, autoencoders), and UI-L integration.
• Custom Solutions : Build and fine-tune domain-specific LLMs or vision-language models for document OCR, field extraction, and medical inference.
• Scalable Infrastructure : Design model-serving pipelines, considering batching, sharding, quantization, and monitoring.
• Continuous Learning : Implement active learning and feedback loops to retrain models based on investigator annotations.
• Performance Analysis : Define and track precision, recall, NGCD, and other metrics; conduct A / B tests and rule simulations.
• Collaboration : Work closely with MLOps, data engineers, and product teams to deploy models in production and iterate rapidly.

Required Qualifications

• 5+ years in machine learning or deep learning engineering.
• Expert in PyTorch or TensorFlow; experience with the Hugging Face ecosystem.
• Strong background in NLP, vision-language models, and graph neural networks.
• Familiarity with model optimization techniques (FT, LoRA, quantization, pruning).
• Solid understanding of MLOps : containerization, monitoring, and CI / CD for ML.
• Excellent communication and documentation skills.

Nice-to-Have

• Experience in healthcare or insurance ML applications.
• Publications in top-tier ML / NLP conferences.
• Knowledge of Bayesian re-ranking, self-supervised learning, or agentic autoML frameworks.

Recruitment Process

• Screening Interview : A brief initial conversation to understand your background and interests.
• Machine Learning Interview : A deep dive into your technical expertise in ML, including model

building and evaluation.

• System Design Interview : Assess your ability to design scalable and maintainable ML systems and

At Qantev, Diversity, Equity and Inclusion is a core principle that drives innovation and success. We are committed to building a global workforce that reflects the rich variety of backgrounds, experiences, and perspectives that make up our world. We hire and embrace applications with no regard to race, ethnic origin, sexual orientation, physical or mental disability, pregnancy, medical condition, gender expression or identity, religion, marital status, age or other non-merit criteria.

This is not a trend for us-it's an integral part of who we are and how we work.

Qantev is proud to foster an environment where everyone has equal access to opportunities, and where each individual can bring their authentic self to the table. We will always strive to welcome applicants of all backgrounds and will remain dedicated to building a team where all voices are heard and all talents are celebrated.

Qantev does not accept unsolicited employment agencies' or headhunter's resumes. We will not pay any third-party that does not have a pre-signed agreement with us. Any unsolicited CVs are deemed to be the property of Qantev and its subsidiaries.