Article - Decision Intelligence
What Makes an AI System Decision-Grade?
A positioning article explaining the difference between model building and end-to-end decision system design.
Draft outlineWriting
Insights, tutorials, and technical notes.
This section is intentionally lightweight for now. The current entries are draft outlines that can become Markdown posts when ready.
Tutorial - Healthcare Decision Support
Predictive Modeling in Healthcare Operations
A practical guide to modeling provider or operational risk while respecting calibration, interpretability, and regulated context.
Draft outlineTechnical Note - GenAI and Retrieval Systems
Evaluation Techniques for LLM and Agentic Systems
A technical note on evaluating LLM systems, RAG workflows, and agents across grounding, task completion, tool use, and failure modes.
Draft outlineTutorial - GenAI and Retrieval Systems
Using Open-Weights LLMs Locally and Effectively
A practical guide to running open-weights models locally with realistic expectations around hardware, quantization, latency, privacy, and quality.
Draft outlineArticle - GenAI and Retrieval Systems
Vector Store Options for RAG: Strengths and Weaknesses
A comparison-oriented overview of vector store choices for RAG systems and the tradeoffs that matter in practice.
Draft outlineArticle - GenAI and Retrieval Systems
LLM Orchestration Layers: When to Use What
A practical comparison of orchestration-layer choices for prompts, chains, tools, agents, retrieval, evaluation, and production workflows.
Draft outlineTechnical Note - GenAI and Retrieval Systems
vLLM vs Hugging Face vs llama.cpp: When to Use Which
A deployment-oriented comparison of common local/open model serving paths and their strengths, weaknesses, and use cases.
Draft outlineTechnical Note - GenAI and Retrieval Systems
RAG Observability and Debugging
A debugging playbook for understanding why RAG systems fail and what to instrument before users lose trust.
Draft outlineArticle - GenAI and Retrieval Systems
RAG Scaling and Budget Optimization
A cost-aware guide to scaling RAG systems without letting retrieval, embedding, context, or generation costs get out of hand.
Draft outline