Description :

We are looking for a Senior MLOps / Machine Learning Engineer with 6+ years of experience to design, build, and scale our next-generation machine learning infrastructure. In this role, you will bridge the gap between Data Science and Core Engineering, ensuring our predictive models move from experimental notebooks to high-throughput, production-grade systems seamlessly.

This isn't a role for standing up low-traffic inference endpoints; we deal with real-world scale. You will work closely with data scientists to optimize distributed training using Ray, orchestrate complex pipelines via Airflow/Composer, and manage scalable deployments on GCP. If you love deep-diving into Python optimization, writing clean Terraform code, and architecture built for high-volume data and traffic, this role is for you.

Requirements :

- Seniority : 6+ years of experience in an MLOps, DevOps, or Data Engineering role with a heavy focus on productionizing machine learning models.

- Python Mastery : Exceptional Python programming skills with a deep understanding of asynchronous programming, performance profiling, and backend frameworks like FastAPI.

- Production ML Scale : Proven track record of deploying and monitoring ML models at scale. You know how to handle high-concurrency traffic, model drifting, and resource optimization.

- Cloud & Data Proficiency : Strong expertise in the GCP ecosystem and writing complex, optimized SQL queries for large datasets.

- Tooling Agility : Comfortable working across a diverse ecosystem of package managers and frameworks, with a keen interest in adopting high-performance tools (like uv).

- Education : Bachelors or Masters degree in Computer Science, Engineering, Mathematics, or a related technical field (or equivalent practical experience).

Job Responsibilities :

- Infrastructure & Automation : Design, provision, and maintain scalable ML infrastructure on GCP using Terraform.

- Scalable Deployment : Architect and deploy high-throughput, low-latency ML inference endpoints capable of handling heavy production traffic.

- Pipeline Orchestration : Build, monitor, and optimize robust data and ML pipelines using GCP Cloud Composer / Apache Airflow.

- Distributed Computing : Implement and scale distributed training and data processing workloads using Ray.

- MLOps Evolution : Transition our current logging and metadata tracking from native GCP tools toward advanced open-source stacks (e.g., MLflow) as our infrastructure matures.

- Collaboration & Standards : Partner with Data Science teams to standardize package management (utilizing uv, Poetry, etc.) and establish best practices for code quality, containerization (Docker), and model reproducibility.