Dheer Gupta

In 2026, all AI generated content sounds the same and it's exhausting to read.

Reviewers default to grading tolerance.

People aren't outsourcing their thinking to LLMs.

Agents in long sessions degrade in a recognizable pattern.

Back in March I shared a command for spawning agent teams.

Building got cheap this year.

Eighteen months was the fast timeline to ship a product.

When an AI session gets me close but not right, I delete everything and start over.

Every layer between customers and the codebase used to protect scarce engineering attention.

The instinct to break work into atomic tickets was right for human teams.

Conway's Law used to show up in JIRA workflows.

I kept seeing the same team dysfunctions in my agent teams that I'd spent years teaching human organizations to fix.

The best product outcomes always came from someone who talked to customers and could also build.

I show the prompt first, then unpack every decision behind it.

I sleep better now because of AI.

Find one subject you know and understand, see if they've encountered the problem in their everyday life.

You may have heard that an idea is only as good as its execution.

Unless you've exhausted all the options that present lower risk, there is no reason to skip ahead to the highest risk play.

Is my innate default sufficiently futuristic?

In 2007 I did a stint on a trading floor.