With over 20 years of experience, our integral network of passionate technologists and pioneering craftsmen deliver cutting-edge technology and game-changing consulting to companies on the brink of transformation. Since 2001, we have grown from a Java company into a full-service digital consulting company with 4500+ professionals working on a worldwide ambition.

We are organized in complementary chapters – teams with a tremendous amount of knowledge and experience within a particular field, such as Agile, DevOps, Data and AI, Cloud, Software Technology, Functional Programming, Low Code, and Microsoft.

We help the world’s top 250 companies and category leaders overcome digital challenges, embrace innovation, adopt new technology, and implement new business models. In addition to high-quality consulting, we also provide offshoring and nearshoring services.

For more details please visit

At Xebia, we put ‘People First’—committed to attracting diverse talent and fostering an inclusive, respectful workplace where everyone is valued for their contributions. We welcome all individuals and evaluate solely on the quality of their work and teamwork.

Role Overview

We are seeking experienced AI/ML Platform ML Engineers to support the development and operationalization of the LIAL product . This role combines strong machine learning expertise with production-grade software engineering to build, scale, and maintain robust ML platform capabilities.

You will be responsible for designing and implementing end-to-end ML workflows—from experimentation and training through to deployment, monitoring, and retraining—ensuring all models are production-ready, reproducible, and governed by best practices.

Key Responsibilities

- Apply strong software engineering discipline to ML development , transforming exploratory notebooks into modular, reusable, and testable Python packages .
- Design and enforce clear interface contracts for ML components to support maintainability and scalability.
- Implement experiment tracking frameworks (e.g., MLflow or Vertex AI Experiments), ensuring:




- Full capture of parameters, metrics, artifacts, and dataset lineage
- Reproducibility of results from a commit hash alone
- Promote best practices for code versioning, testing, and documentation across ML workflows.

- Training, Evaluation & Hyperparameter Optimization

- Design and implement distributed training pipelines (multi-GPU / multi-node), ensuring:
- Robust checkpointing
- Fault tolerance and recoverability

- Develop standardized evaluation templates that include:

- Core performance metrics
- Bias and fairness assessments
- Shadow-mode testing against baseline models

- Move beyond simple validation by ensuring models are evaluated under realistic production scenarios .

- Model Packaging, Serving & Deployment

- Build and maintain standardized model packaging templates , including:
- TensorFlow SavedModel
- TorchScript
- ONNX

- Create versioned, production-ready serving containers and publish them to registries (e.g., Artifact Registry).
- Canary releases with traffic splitting

- Safe rollback procedures
- Both real-time (online) and batch inference use cases
- Leverage platforms such as Vertex AI Endpoints to operationalize model serving.

- Production Monitoring & Retraining

- Design templates covering the entire post-deployment lifecycle , including:
- Prediction quality monitoring
- Clearly defined alert thresholds
- Build and maintain automated retraining pipelines triggered by monitoring signals.

- Define and enforce model lifecycle governance , including:

- Operational runbooks
- Ensure no model operates in production without observability and traceability .

- Ways of Working & Engagement

- Work independently and autonomously ,



owning deliverables end-to-end.
- Apply a security-first mindset in all platform and ML engineering activities.
- Demonstrate strong:
- Planning and prioritization skills
- Communication and stakeholder engagement
- Reporting and documentation discipline
- Collaborate effectively within cross-functional teams including product, data, and platform engineering.

Required Experience & Skills

- Proven experience as an ML Engineer in production environments (not purely research-focused).
- Strong proficiency in Python and modern ML frameworks (TensorFlow, PyTorch).
- Hands-on experience with:
- ML lifecycle tooling (MLflow, Vertex AI, or equivalent)
- Distributed training and scalable compute environments
- Containerization (Docker) and deployment pipelines
- Experience with cloud-native ML platforms , preferably Google Cloud / Vertex AI .

- Solid understanding of:

- Model evaluation beyond accuracy (fairness, robustness, monitoring)
- CI/CD for ML systems (MLOps practices)
- Familiarity with artifact management and version control systems .

Nice to Have

- Experience building enterprise AI/ML platforms supporting multiple teams/products
- Knowledge of data governance, lineage, and compliance frameworks
- Exposure to high-scale ML systems and real-time inference architectures
- Experience implementing automated retraining and adaptive learning systems

Engagement Model

- Focus: Delivery of AI/ML Platform Engineering capabilities to support LIAL product development
- Working Style: Autonomous delivery with structured reporting and stakeholder alignment

Success Criteria

- Reproducible ML pipelines with full traceability
- Production-grade deployment patterns with zero-downtime releases
- Robust monitoring and automated retraining pipelines in place
- Standardized templates enabling scalable ML development across teams

Compensation

Salary Range: €68,000 – €82,000 gross per year, depending on experience, skills, and overall fit for the role.

#J-18808-Ljbffr

📌 AI/ML Platform Cloud Infrastructure Engineer - LIAL product. (España)
🏢 Xebia
📍 España