AI Engineering
Architecture, roles, and engineering practices for AI-based product development.
Articles
LLM-Only vs RAG GenAI Apps: Architecture Impacts Cost, Quality, Trust
Part 2 of the GenAI engineering series. When to use LLM-only, when to use RAG, and how poor architecture decisions increase token cost, latency, and reliability issues.
Understanding GenAI Hardware: CPU, GPU, NPU, Inference, and Model Serving
A practical guide to how GenAI workloads run on hardware, when to use CPU, GPU, or NPU, and how inference, embeddings, RAG, and model serving fit together.
Getting Started with Apple MLX for Local AI and LLM App Development
Learn how to set up Apple MLX and mlx-lm on Apple Silicon, run local LLM inference, and expose model generation with a FastAPI API for practical AI app development.
Why AI Product Development Still Needs Architecture Thinking
AI can generate code quickly, but architecture decisions around trade-offs, security, scalability, and maintainability still require human judgment.
How to Reduce AI Coding Tool Costs with Better Prompting and Context Engineering
AI coding costs drop when teams use better specs, smaller tasks, reusable context, test-first workflows, and fewer broad exploratory prompts.
AI-Based Product Development: Roles Still Matter as AI Accelerates Execution
AI can accelerate product delivery, but product, architecture, engineering, QA, DevOps, security, support, and operations still need clear ownership.