Machine Learning Engineer

Job description

**Job Title: Machine Learning Engineer – Tribe AI** **Department** : ISPL ITG Fortis ADM **About Business Line/Function** BNP Paribas Fortis is a key player on the Belgian market and a driver of economic growth. Within the entity The Tribe Artificial Intelligence aims to deliver an efficient and seamless banking experience for our customers and empower our employees through AI. Our expert team of approximately 90 experts includes Data Scientists, Machine Learning Engineers, Business Analysts, Scrum Masters, Product Owners and Managers. We develop AI applications to address various business challenges, including virtual assistants to enhance client experiences, advanced AI tools to support employee efficiency and creativity, and automation to optimize internal processes. **Position Purpose** You will drive the industrialization of AI solutions, ensuring they are reliable, scalable, and aligned with technical best practices. You will act as the key bridge between Data Scientists and IT Production team, translating business needs into production-ready ML pipelines while anticipating technological advancements. As part of the Machine Learning Engineers Chapter, you will collaborate closely with Brussels-based teams, maintaining seamless Data sourcing for the Tribe’s AI use cases, and help the teams maintain a high level of obsolescence management and security in the python environments. Your role is both strategic and operational, transforming prototypes into high-performance AI services while fostering excellence in MLOps, code quality, and cross-team collaboration. **Responsibilities** Direct Responsibilities - As a ML Engineer you will be part of a team that is responsible of the following operational activities: - Design, Maintain and optimize data sourcing pipelines using ETL, CFT, Denodo (data virtualization), and Airflow to ingest, transform, and expose data for AI/ML use cases. Ensure seamless integration of new data sources (internal/external APIs, databases, or streaming platforms) while adhering to data governance and latency requirements. - Maintain Python environments by proactively auditing dependencies, upgrading obsolete libraries, and enforcing version compatibility across development, testing, and production. Document and communicate changes to minimize disruption. - Enforce Vulnerability management in production code by: - Conducting regular security scans and patching critical vulnerabilities in pipelines, APIs, and dependencies. - Implementing secure coding practices and collaborating with cybersecurity teams to mitigate risks. - Automating compliance checks in CI/CD pipelines to block vulnerable code from deployment. - Understand & support CI/CD workflows (Jenkins, GitLab CI/CD) for containerized ML models (Docker/Kubernetes), ensuring seamless deployment, versioning, and rollback capabilities. - Troubleshoot and resolve complex incidents in QA/Production, ensuring minimal downtime and continuous improvement of AI services. - Collaborate with Data Scientists and PROD IT teams to define production-ready architectures, balancing technical feasibility with business requirements (real-time responses, high-volume processing). - Promote Software Engineering best practices— code quality, security, logging,… —within your squad. - Stay ahead of AI/ML advancements (LLMs, Agentic AI) and propose innovative solutions to optimize workflows and reduce time-to-market. **Technical & Behavioral Competencies** Nice-to-have: - Mandatory : Expert - >4 years of professional experience in Python Programming (OOP, decorators, code quality & security, performance optimization) - Python environment building : strong uv skills, pip, mamba, micromamba, - ML engineering: MLOps, model versioning, deployment. - Containerization & orchestration: Docker, Kubernetes (scaling, resource management). - CI/CD pipelines: Jenkins, GitLab CI/CD (advanced workflows, artifact management). - Linux/Cloud infrastructure: Bash scripting, system administration, troubleshooting. - Database systems: PostgreSQL (query optimization, schema design). - Monitoring & incident management: Advanced logging & analysis, debugging complex issues. - Denodo Platform Proficiency – Ability to configure, query, and optimize virtual data layers using Denodo’s data virtualization tools, including creating logical views, data services, and API integrations. - Strong SQL skills to write efficient queries in Denodo’s VQL (Virtual Query Language) and optimize data retrieval for AI model training and validation. - Data Virtualization & Integration – Experience in connecting disparate data sources (e.g., databases, APIs, ETL pipelines) via Denodo to enable seamless data exploration for AI/ML workflows. - Airflow DAG Development & Orchestration – Ability to design, implement, and maintain scalable Directed Acyclic Graphs (DAGs) for AI/ML pipelines, including task dependencies, retries, and dynamic workflow generation - Airflow Integration & Optimization – Experience in: - Connecting Airflow to data platforms (Denodo, PostgreSQL, S3, etc.) and ML tools (e.g., MLflow, Kubeflow) via hooks, custom operators, or APIs. - Optimizing performance through parallelism tuning, executor selection (Celery/Kubernetes), and efficient XComs/artifact handling for large-scale workflows. - API design (Django framework). - Distributed ML systems (Spark, Hadoop) and ETL/ELT pipelines (Airflow). - Model compression/optimization (quantization, pruning). - Data visualization (Kakfa, Kibana, ELK, Grafana) for monitoring dashboards.Specific Qualifications: - Minimum 6 years of experience along with Masters Degree - in IT, Computer related field or equivalent experience. - Prior experience in similar domain. **Other Skills** **Skills Referential** **(** **Required knowledge, skills and abilities)** Your English level is top-notch to create bonds & collaborate with your peers. - Agile methodology has no secret for you.