Senior ML Ops Engineer - Dating App - GCP, Vertex AI (M/F/X)

Présentation de l'entreprise

Join us and experience something truly unique in your career.

Présentation du poste

70–80 K€ gross per year + ~5% bonus + profit sharing

Paris (hybrid, 3 days on-site / week)

Remote: 2 days / week after onboarding

English: fluent / French: nice to have

5+ years in ML Engineering / MLOps / Software Engineering with strong ML in production

Join a European Dating App leader where data and ML are at the heart of the product. As an Senior ML Ops Engineer, you will be the reference MLOps engineer in Paris, working hand-in-hand with a senior MLOps / ML Engineering team in North America and a Data Science team in Paris. You will own the full lifecycle of ML models in production on a modern GCP stack, from training pipelines to monitoring and incident handling.

What you will do

• Put ML models into production end to end: training, deployment, retraining, rollbacks.
• Design, maintain and improve MLOps pipelines (CI/CD, data flows, orchestration).
• Own reliability, performance and availability of ML systems in production.
• Implement and operate monitoring, logging and alerting for all ML services.
• Handle production incidents related to ML models and drive post-mortems.
• Support Data Scientists t o industrialize their models and experiments.
• S hare best practices and help shape the ML architecture and standards.
• Act as the bridge between the Data Science team in Paris and the MLOps / platform team in North America.

Stack & environment :

Language: Python.

Cloud : GCP (BigQuery, Cloud Run, IAM, service accounts).

ML services: Vertex AI (training, endpoints, pipelines).

Deployment: containerized services on Cloud Run, Vertex AI Endpoints.

CI/CD: automated pipelines, GitHub-based workflows.

IaC: Terraform for GCP infrastructure.

Monitoring & observability: metrics, logs, alerts, Grafana.

Data: BigQuery, event-driven components with Kafka, some legacy Spark/Scala (being phased out).

Organisation: centralized Data Hub (Data Science, ML/Data Engineering, BI, DBAs), international teams across Europe, Canada and US.

Profil recherché

Who we’re looking for

• 5+ years as ML Engineer / MLOps / Software Engineer with strong ML production experience .
• Proven track record putting ML models into production and running them reliably.
• Solid production mindset: incidents, SLAs, monitoring, technical debt do not scare you.
• Strong skills in Python, Docker, CI/CD, Terraform, GCP (Vertex AI, Cloud Run, BigQuery).
• Comfortable working closely with Data Scientists and platform teams.
• Very good communication in English; able to collaborate daily with North American teams.
• Autonomous, rigorous, pragmatic, comfortable as a technical reference without direct reports.

Nice to have: Kafka, Spark, ElasticSearch, Grafana, A/B testing frameworks, BI tools.

Package & conditions :

• Salary: 70–80 K€ gross per year + ~5% bonus.
• Additional: profit sharing (intéressement and participation) + benefits (lunch vouchers, healthcare, mobility, fitness, etc.).
• Contract: full-time permanent position.
• Location: Paris
• Remote: 2 days remote / week

Why this role is attractive :

• High-impact ML : work on matching, coaching, trust & safety, business scoring for millions of users.
• Modern stack: GCP, Vertex AI, Cloud Run, BigQuery, Terraform, Grafana; low legacy.
• International culture: daily collaboration with Canada and US, multi-brand environment.
• Strong learning environment: e-learning, conferences, knowledge-sharing, hackathons.

Hiring process :

Step 1 – Recruiter screen (video, 30–45 min): career path, motivations, compensation, logistics.

Step 2 – Hiring Manager interview (video, 45 min): role understanding, collaboration, communication.

Step 3 – Technical interview (video, 60 min): Python coding (algorithms, basic data processing) + technical Q&A with Data Scientists.

Step 4 – Technical deep-dive (video, 90 min, full English): live TensorFlow exercises and discussion with a senior ML engineer.

Step 5 – Final interview (preferably on-site, 60 min): meeting with Engineering leadership and People team