Senior MLOps Engineer

We are looking for a Senior MLOps Engineer to add to our Data team in Barcelona.

Join the User Acquisition MLOps team that builds the core ML backbone of Appodeal. We design and run the systems that power real-time predictions at massive scale, where every millisecond matters. Our work directly drives thousands of bidding operations each second, supporting a high‑throughput, low‑latency inference platform used across the company.

You will help shape and evolve a modern ML platform built around Databricks, MLflow, Unity Catalog, GitHub Actions, and deep integrations with our Predictor service. This means solving real engineering challenges: scaling training pipelines, optimizing GPU workloads, ensuring safe and automated model promotions, and keeping online inference reliable under extreme load.

You will collaborate daily with Data Science, Data Engineering, Runtime, DevOps, and Product, and you will fully own the systems and projects you touch. This is a role for engineers who like autonomy, impact, and building things that matter.

Responsibilities:

• Design, operate, and improve the ML platform, including Databricks workflows for training, MLflow and Unity Catalog for model management, CI/CD pipelines using GitHub Actions and Databricks Asset Bundles, and integration with the runtime inference layer.
• Support data scientists in developing and maintaining end-to-end training pipelines, covering data processing, feature transformations, ML model training on GPUs, and distributed training setups to ensure efficient and scalable workflows.
• Ensure reliable and safe model inference by validating models before promotion, checking metadata compatibility, monitoring latency, and supporting model evaluation flows.
• Maintain strong ML observability, including online and offline monitoring, data and feature parity checks, and automated validation of model performance in production.
• Develop internal tools and APIs that enable DS teams and account managers to experiment, validate, and promote models efficiently while reducing operational friction.

Requirements:

• 3+ years of experience in MLOps, ML Platform Engineering, or Backend roles supporting production ML systems.
• Strong Python and software engineering skills, with experience building and maintaining APIs (FastAPI, Flask, or similar).
• Hands-on experience with DNNs using PyTorch or other modern deep learning frameworks, including GPU training and optimizing hardware utilization.
• Solid understanding of distributed computing and data processing using Apache Spark or similar frameworks.
• Experience with Databricks (Jobs, MLflow, Delta, UC) or an equivalent cloud-based ML platform.
• Strong understanding of the ML lifecycle: data ingestion, experimentation, model packaging, deployment, and monitoring.
• Familiarity with real-time model inference systems and ensuring compatibility with downstream services.
• Experience with data validation libraries (Pydantic or similar) used within ML and API workflows.
• Good understanding of Docker containers, Kubernetes, and basic networking concepts.
• Experience creating and maintaining CI/CD pipelines (GitHub Actions or similar).
• Experience with orchestration tools (Databricks Jobs, Dagster, Airflow or similar).
• Demonstrated ability to use modern AI tools (Claude, ChatGPT, Gemini) to improve development speed and quality.
• Proactive mindset with a strong sense of ownership, consistently pushing work forward and anticipating next steps without needing close direction.
• Strong English communication skills.

Nice to have:

• Exposure to very high-load, low-latency ML inference environments.
• Knowledge of distributed or large-scale ML training frameworks and multi-GPU setups.
• Familiarity with Databricks Asset Bundles.
• Understanding of ONNX or other portable model serialization formats.
• Ability to work with Rust programming language.
• Familiarity with transformer architectures, encoder-decoder models, or sequence modeling approaches.
• Exposure to data engineering concepts, such as building scalable data pipelines, understanding CDC patterns, or working with structured and unstructured datasets.
• Experience with MinIO or other S3-compatible storage systems.
• Understanding of networking protocols such as TCP, HTTP, and gRPC, including their limitations in low-latency scenarios.
• Confidence navigating the Atlassian ecosystem (Jira, Confluence, Compass).

This position supposes relocation to Barcelona and work from our office in the city center.

Why Join Us:

• Opportunity to work on cutting-edge projects with a global impact in the mobile app industry.
• A collaborative and inclusive work culture that values innovation and continuous learning.
• Competitive salary, flexible work arrangements, and a comprehensive benefits package.
• Professional development opportunities and career growth prospects within a fast-growing company.