Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Duration 14 hours
Course Outline
Introduction to Devstral and Mistral Models
- Overview of Mistral’s open-source models
- Apache-2.0 licensing and enterprise adoption
- The role of Devstral in coding and agentic workflows
Self-Hosting Mistral and Devstral Models
- Preparing environments and selecting infrastructure
- Containerisation and deployment using Docker/Kubernetes
- Scalability considerations for production usage
Fine-Tuning Techniques
- Supervised fine-tuning versus parameter-efficient tuning
- Dataset preparation and cleaning processes
- Examples of domain-specific customisation
Model Ops and Versioning
- Best practices for model lifecycle management
- Strategies for model versioning and rollback
- CI/CD pipelines specifically for ML models
Governance and Compliance
- Security aspects of open-source deployment
- Monitoring and auditability within enterprise contexts
- Compliance frameworks and responsible AI practices
Monitoring and Observability
- Tracking model drift and accuracy degradation
- Instrumenting inference performance
- Establishing alerting and response workflows
Case Studies and Best Practices
- Industry use cases for Mistral and Devstral adoption
- Balancing cost, performance, and control
- Lessons derived from open-source Model Ops
Summary and Next Steps
Requirements
- A solid understanding of machine learning workflows
- Proficiency with Python-based ML frameworks
- Familiarity with containerisation and deployment environments
Audience
- ML engineers
- Data platform teams
- Research engineers
