This repository documents my structured journey toward becoming an Applied AI / LLM Systems Engineer. It contains all study notes, architectural principles, design rules, experiments, and implementation artifacts across every phase of the roadmap, from systems mindset and probabilistic reasoning to production-grade AI services with observability, evaluation, and deployment discipline.
The focus is system-first engineering, not trend-driven experimentation. Every concept is tied to concrete implementation, failure modes, cost considerations, and production reliability.
The repository covers:
Phase 0: Systems mindset, AI design rules, deterministic vs probabilistic thinking
Phase 1–2: RAG engineering, structured LLM usage, failure-aware AI systems
Phase 3: Strongly-typed backend systems (.NET Core)
Phase 4: Model engineering, ONNX interoperability
Phase 5: DevOps, reliability, observability
Phase 6: LLM evaluation, gold datasets, regression pipelines
All learning is documented through markdown notes, architectural breakdowns, and implementation-backed experiments.
Goal: Build AI systems that are controllable, observable, cost-aware, and production-ready.