Training material
LLM platforms hands-on
Site reliability engineering for applied AI
Run, observe, secure and deliver Mistral models on Linux and Kubernetes at customer sites – trained in your own lab.
- Linux
- Python
- Go
- Mistral
- Mandanten
- Metriken
- SLOs
- Security
Training plan
- Foreword
- Day 0: Build the training ground
- Day 1: Linux internals
- Day 2: Network debugging
- Day 3: Python for SRE – llmctl
- Day 4: Go for SRE – llm-gateway
- Day 5: Mistral models on a Linux VM
- Day 6: Multi-tenant Kubernetes
- Day 7: Model serving on Kubernetes
- Day 8: Observability I – metrics
- Day 9: Observability II – logs and traces
- Day 10: SLOs, Alerting and On-Call
- Day 11: Incident Response and Post-Mortem
- Day 12: Infrastructure as Code
- Day 13: Security – Secure SDLC and Supply Chain
- Day 14: Hybrid and On-Prem at the Customer
- Day 15: Competition – Disaster Drills
- Appendix A: Cheat Sheet
- Appendix B: Final Exam
- Appendix C: Interview Preparation Applied AI / SRE