A progressive series of AI systems — from transparent local pipelines to production-grade cloud deployments.
These projects are built with incremental learning in mind. Each one introduces new layers of complexity — starting from simple, fully transparent AI systems that run locally, and progressing toward enterprise-grade architectures deployable on AWS, Azure, or across both.
The goal is to build solid conceptual foundations at every step. Rather than jumping straight to managed services and abstractions, each project exposes the mechanics underneath: how retrieval works, how infrastructure is composed, how security and governance are applied at scale. Solutions are designed to run locally for development and exploration, and to scale to cloud when production patterns are needed.
| Project | Domain | Status | Deployment |
|---|---|---|---|
| Simple RAG Demo | Generative AI · RAG · Local LLM |  |
Local |
| Simple RAG Demo v2 | Generative AI · RAG · Hybrid Retrieval · Observability | |
Local |
| Simple RAG Demo v3 | Generative AI · RAG · Groq LLM · Langfuse Observability | |
Local + Cloud LLM |
| AI Platform Operations | DevOps · Platform Engineering · Security · Governance | |
AWS |
Enterprise-AI-Solutions/
├── simple-rag-demo/ # Local RAG pipeline — LangChain, FAISS, Ollama, Streamlit
├── simple-rag-demo-v2/ # Enhanced RAG — BGE embeddings, hybrid retrieval, observability layer
├── simple-rag-demo-v3/ # Groq LLM + Langfuse observability — traces, spans, generation metrics
└── ai-platform-operations/ # AWS operational framework — Terraform, Bedrock, SageMaker
MIT — see individual project directories for full license text.