MLOps

We are looking for an MLOPS
Required Skills:

• Google Cloud Platform (GCP): Solid and demonstrable experience with Vertex AI (Pipelines, Model Registry, Workbench, and model deployment) and BigQuery (large-scale query optimization, distributed processing, and external connectors).

• Software Engineering / Python: Advanced level in Python focused on developing modular, decoupled, testable code prepared for concurrent execution. Experience applying engineering best practices (Git, automated testing, code reviews, and dependency management).

• Machine Learning for large data volumes: Experience developing predictive models with algorithms such as LightGBM, XGBoost, CatBoost, or similar, working with datasets of millions of records and optimizing training and inference times.

• Model Industrialization (MLOps): Experience taking models from experimentation environments (Notebooks/Workbench) to production through automated pipelines, model versioning, reproducible deployments, and monitoring.

• DevOps applied to Machine Learning: Experience with Docker, CI/CD (GitHub Actions, GitLab CI, or similar), deployment automation, and environment management for Machine Learning solutions.

• Data Engineering for Machine Learning: Experience building efficient data pipelines for training and inference, optimization of ETL/ELT processes, and management of large data volumes.

• Forecasting / Time Series: Experience in demand forecasting projects with high cardinality, handling complex seasonality, promotions, events, and cold-start strategies for new products or entities.

• Deep Learning for Forecasting: Knowledge of modern architectures for time series such as Temporal Fusion Transformer (TFT), TiDE, PatchTST, TimesFM, Chronos, or Transformer-based models.

• Generative AI (GenAI / LLMOps): Experience in the development or deployment of solutions based on foundation models (LLMs), including RAG, agents, model evaluation, prompt engineering, observability, and lifecycle management of Generative AI applications.

• Model optimization and performance: Experience in hyperparameter optimization techniques (Optuna, Ray Tune, or similar), training parallelization, inference optimization, and reduction of computational costs.