Workflow

Databricks Integration Deep Dive: Teaching Aye Chat to Talk to Your Endpoint

A pragmatic walkthrough of Aye Chat’s Databricks model plugin: how it activates, what it sends (OpenAI-style), how it stores history without hoarding your repo, and how it extracts JSON from the model’s inevitable ‘creative writing.’

Stop Re-Explaining Yourself to the AI: The Case for Behavior Presets

You keep writing the same preamble in every prompt. The AI keeps ignoring half of it. There’s a better pattern: repo-local behavior presets that make your assistant consistent without making your prompts enormous.

Why I Use Different AI Models for Planning, Reviewing, and Coding

A walkthrough of using GPT-5.2 for planning, Claude Opus 4.6 for validation and implementation - and why switching models at each stage caught bugs I would have shipped otherwise.

Claude Code Asks Nicely. Aye Chat Defaults to Action.

Claude Code’s ad celebrates ‘explain first, wait for approval.’ Aye Chat bets on a different UX: act first, snapshot everything, and make undo so cheap you stop negotiating with the model.

Claude Code Asks Nicely. Aye Chat Swings the Hammer (with an Undo Cord)

Claude Code’s ad celebrates ‘explain first, wait for approval.’ Aye Chat bets on a different UX: act first, snapshot everything, and make undo so cheap you stop negotiating with the model.

The Day I Stopped Babysitting an AI: Building the Optimistic Workflow

The story of how frustration with approval-seeking AI assistants led to Aye Chat’s ‘optimistic workflow’ - where the AI acts first and you review after, backed by an instant undo button.