Jonathan Barazany - AI Agents & Engineering Leadership

Eight months of running automated AI code review on my own projects.

For months I kept correcting Claude the same ways.

Everyone's complaining about the Claude Code quota.

A few months ago I described context engineering from the outside.

We spent weeks in governance spirals over n8n and Workato.

I was stuck at 0.58 AUC for three weeks.

After our Claude Code pilot hit 65% AI-written code, the most common question was: how did you actually measure that?

A pilot group of 30-40 engineers set a 40% AI-written code target.

It took a month to connect Snowflake to Claude.

A development team moves so fast that Product can't keep up.

Cursor 2.0 reimagines code editors as orchestration platforms.

A small, fast, affordable model that matches Sonnet 4-level capabilities.

GLM-4.6 is gaining attention as a capable, affordable alternative to Claude, but after nine months with Sonnet models, it's solid progress without redefining the frontier. (2 min read)

Behind every long-running AI agent, there's quiet, invisible engineering that keeps the model from forgetting who it is and what it's doing. (6 min read)

A week of rapid-fire AI releases changed the game.

From manual pain to automated precision: how I used MCP tools to turn the nightmare of debugging complex Kusto queries into structured, data-driven performance reports. (4 min read)

How I turned a painful Spark debugging cycle into a tight feedback loop using Databricks CLI, DuckDB, and local iteration.

A production mystery that changed everything I thought I knew about Kubernetes resource management.