Machine Learning
Survival Analysis for Data Drift and ML Reliability
July 7, 2026
Survival Analysis for Data Drift and ML Reliability
Introduction systems rarely fail in a single moment. Their performance changes gradually as data distributions shift, calibration drifts, or new…
What's new in managed agents in the Gemini API
July 7, 2026
What's new in managed agents in the Gemini API
Today we're announcing new Managed Agents capabilities in the Gemini API, including background execution, remote MCP server integration, custom callbacks…
How to Run End-to-End Tests with Claude Code
July 6, 2026
How to Run End-to-End Tests with Claude Code
like the Claude Code can be used for a variety of tasks. The first task you might think of is…
Validating the RAG Answer Before the User Sees It: Spans, Quotes, and the Feedback Loop
July 6, 2026
Validating the RAG Answer Before the User Sees It: Spans, Quotes, and the Feedback Loop
the generation brick of Enterprise Document Intelligence, a series that builds an enterprise RAG system from four bricks: document parsing,…
Stop Ranking Agent Configs by Average Score
July 6, 2026
Stop Ranking Agent Configs by Average Score
Stop Treating Agent Eval Like a Leaderboard eventually hits the same question: which version should I actually ship? You change…
Assemble Each RAG Generation Prompt from a Base Prompt Plus the Rules Each Question Needs
July 5, 2026
Assemble Each RAG Generation Prompt from a Base Prompt Plus the Rules Each Question Needs
part of the generation brick of Enterprise Document Intelligence, a series that builds an enterprise RAG system from four bricks:…
PANet Paper Walkthrough: When Feature Pyramids Go Bottom-Up
July 5, 2026
PANet Paper Walkthrough: When Feature Pyramids Go Bottom-Up
I wrote about the FPN (Feature Pyramid Network) architecture [1], which is one of the most influential necks we can…
Setting Your Own Great Language Model
July 4, 2026
Setting Your Own Great Language Model
: Frontier AI models are increasingly at risk of being trapped behind strict export controls or mounting API costs. As…
Stop Returning Text from RAG: The Typed Answer Contract That Prevents Hallucination
July 4, 2026
Stop Returning Text from RAG: The Typed Answer Contract That Prevents Hallucination
brick of Enterprise Document Intelligence, a series that builds an enterprise RAG system from four bricks: document parsing, question parsing,…
AI Agents Explained: What Is the React Loop and How Does It Work?
July 3, 2026
AI Agents Explained: What Is the React Loop and How Does It Work?
In my last post,. Calling a Tool is a method that allows an AI model to decide which function to…