AI / ML Engineer

At Ardanis, we’re seeking an AI / ML Engineering Specialist with solid experience in enterprise-grade machine learning systems, Cloudera ML stack integration, and RAG (Retrieval-Augmented Generation) pipelines. The role involves end-to-end ownership of ML infrastructure, from data ingestion and feature engineering to model deployment, monitoring, and lifecycle automation. You’ll work closely with data engineers, ML scientists, and platform architects to operationalize AI workloads within distributed environments, ensuring performance, scalability, and reproducibility.

Responsibilities :

• Design and implement production-ready ML pipelines on the Cloudera Data Platform (CDP), integrating Cloudera Data Engineering, Data Science Workbench, and Machine Learning components.
• Develop and maintain RAG architectures, combining embedding models, vector databases, and LLM inference layers for enterprise-scale retrieval and reasoning.
• Integrate vector stores such as FAISS, Milvus, Pinecone, or ChromaDB into existing data pipelines and Cloudera ML workflows.
• Implement feature stores, model registries, and CI / CD pipelines for automated deployment and retraining using MLflow, Airflow, and Kubernetes.
• Optimize model inference latency and resource utilization in distributed environments (Spark, YARN, K8s).
• Develop REST / gRPC APIs and microservices to serve models and RAG endpoints.
• Monitor model drift, retraining triggers, and lineage using observability tools and metadata tracking systems.
• Collaborate with data platform engineers to ensure compliance with data governance, lineage, and access control standards in Cloudera.

Requirements

You’ll be joining a cross-functional engineering team that designs, deploys, and maintains production-grade AI infrastructure. The environment is highly collaborative, code-driven, and version-controlled, with emphasis on reproducibility, data lineage, and system observability. We expect strong ownership, deep understanding of MLOps principles, and the ability to reason about trade-offs in performance, scalability, and model lifecycle management.