Model Context Protocol guide
How connected AI tools use external data, tool calls, and services.
AI Engineering
Practical notes on building AI systems that survive real workflows: LLM applications, agentic automation, evaluation, retrieval, MLOps, deployment, observability, and the data platforms underneath them.
LLM Systems and Agents
Autonomous AI workflows with n8n
A technical guide to agentic automation systems and workflow design.
Enterprise document transformation agent
Building document processing workflows with LlamaReport and LlamaCloud.
Uncertainty quantification for LLMs
A hands-on guide for understanding confidence and uncertainty in LLM output.
Production ML and MLOps
MLOps essential skills
Model management and deployment foundations for production ML work.
Docker for machine learning
Using Docker for reproducible data science and machine learning environments.
Ray for distributed ML
Distributed compute patterns for machine learning workloads.
Trustworthy machine learning
Notes on transparency, fairness, privacy, robustness, and accountability.
AI Infrastructure
Databricks lakehouse guide
Delta Lake, MLflow, Unity Catalog, and data platform implementation notes.
Kubernetes guide
Container orchestration concepts and implementation patterns.
Deep learning engineering with JAX
High-performance machine learning and numerical computing patterns.
Building and publishing a Python package
Modern Python packaging, testing, documentation, and release automation.
Editorial Scope
This column is for the engineering layer between AI demos and dependable products: system design, integration, evaluation, reliability, data quality, deployment, and developer workflow. General ML theory still appears in the main blog, while this page collects the posts closest to shipping AI systems.