Machine Learning in Production

Source: https://edu.kyrylai.com/courses/ml-in-production

Elevate your career in machine learning to the next level by mastering the complete end-to-end process—from setting up infrastructure to deploying models.This course is your pathway to becoming a fully-fledged ML specialist. In just 8 weeks, you will transition from a data scientist focused on models to an engineer capable of managing the entire ML lifecycle. You can choose to learn through live streaming (cohort) or take the course at your own pace. Course Program:

• Week 1: Setting up and managing Docker, Kubernetes, and CI/CD pipelines
• Week 2: Advanced approaches to data storage, processing, versioning, and annotation, as well as Retrieval-Augmented Generation (RAG)
• Week 3: Setting up, conducting, and optimizing experiments to achieve maximum model efficiency
• Week 4: Automation and orchestration of ML processes using Dagster, Kubeflow, and Airflow
• Weeks 5–6: Deployment, scaling, and maintenance of models, including working with large language models (LLM)
• Week 7: Monitoring, maintaining, and managing the quality of models, including tracking data drift and monitoring LLMs
• Week 8: Vendor selection and platform integration (AWS SageMaker, GCP Vertex AI), as well as analysis of current trends

Final Project:

Completion of a full end-to-end ML project with a subsequent presentation, where you will apply all acquired knowledge in practice. Learning Outcomes:

• Capstone Project: An implemented ML project demonstrating the ability to solve real-world problems from start to finish
• Design Document: Detailed architecture and description of the ML system refined during the course
• Reusable Code Templates: A set of practical solutions and templates for future projects

Password: cms.ddpanda.org