How ai-infra-curriculum/ai-infra-performance-learning works
This is a content-first curriculum repository, not a software runtime: learners follow an advanced AI/ML performance-engineering course from GPU fundamentals through CUDA and profiling into model/transformer optimization, distributed inference, and production deployment. Projects provide applied work, with assessments and learning resources as supporting branches. The roadmap and module sequence are documented in the README and tree; sampled executable sources are exercise starters and do not establish a shared runtime or grading service.
Main entry points: mod-004-transformer-optimization/README.md Transformer Optimization · mod-005-model-compression/README.md Model Compression · mod-006-distributed-inference/README.md Distributed Inference · README.md Curriculum Overview
Components 16
Core Learning Path
- Curriculum Overview
README.md - GPU Fundamentals
modules/mod-001-gpu-fundamentals/README.md - CUDA Programming
modules/mod-002-cuda-programming/README.md - Performance Profiling
modules/mod-003-performance-profiling/README.md
Optimization Topics
- Transformer Optimization
modules/mod-004-transformer-optimization/README.md - Model Compression
modules/mod-005-model-compression/README.md
Inference And Deployment
- Distributed Inference
modules/mod-006-distributed-inference/README.md - Production Deployment
modules/mod-007-production-deployment/README.md - Advanced Topics
modules/mod-008-advanced-topics/README.md
Applied Learning
- Compression Project
projects/project-01-model-optimization/README.md - GPU Kernels Project
projects/project-02-gpu-optimization/README.md - Inference Project
projects/project-03-distributed-inference/README.md - Quizzes And Exams
assessments/README.md - Learning Resources
resources/README.md - Community Support
community/README.md
Other components
- Learner
Connections 15
- Learner → Curriculum Overview: begins at
- Curriculum Overview → GPU Fundamentals: starts with
- GPU Fundamentals → CUDA Programming: advances to
- CUDA Programming → Performance Profiling: advances to
- Performance Profiling → Model Compression: continues to
- Model Compression → Distributed Inference: continues to
- Distributed Inference → Production Deployment: continues to
- Distributed Inference → Advanced Topics: branches to
- Model Compression → Transformer Optimization: branches to
- Production Deployment → Inference Project: applies in
- Transformer Optimization → Compression Project: applies in
- Transformer Optimization → GPU Kernels Project: applies in
- Learner → Quizzes And Exams: completes
- Learner → Learning Resources: consults
- Learner → Community Support: seeks support