Open main menu
AboutAdd-ons
Partner with us →
Lola

Lola

Lola is the founder of Lunch Pail Labs. She enjoys discussing product, app marketplaces, and running a business. Feel free to connect with her on Twitter or LinkedIn.

The Head Chef Model of AI Collaboration
Build
•

The Head Chef Model of AI Collaboration

The Head Chef Model reframes how to work with AI. Not as a manager, intern, or copilot, but as a hands-on creative lead. You stay in the kitchen, guiding taste, timing, and vision while AI handles execution. The more fluent you are in your craft, the better you can shape what AI produces.

Speed is a commodity
Company
•

Speed is a commodity

When Facebook shouted "move fast and break things," speed felt like a secret weapon only a few could wield. Today AI turns speed into a commodity. The real advantage isn't moving faster, it's building systems that let AI sprint while you focus on the deeper processes that sustain quality work.

DRY Applies to Agents Too
Build
•

DRY Applies to Agents Too

I used to manage all my agent instructions in one file. It worked at first, but it didn’t scale. Splitting them into modular files by domain made my workflows faster, easier to maintain, and way less fragile. Here’s how I made the shift and what I’m experimenting with next.

Make Background Coding Agents Actually Work For You
Build
•

Make Background Coding Agents Actually Work For You

Background coding agents are finally affordable and accessible, but most still produce low-quality output. The issue isn’t the models — it’s the missing context. In this post, I share how adding a simple swarm.notes.md file to every repo turned my AI agents into actual contributors, not cleanup jobs.

The hidden lever that makes AI agents actually useful
Build
•

The hidden lever that makes AI agents actually useful

After building my own AI swarm for over a year, I realized the most important variable is the trigger. In this post, I break down how to design agent execution loops that actually get things done, without needing constant attention.

Your instructions are probably a mess. Mine were too. XML fixed it.
Build
•

Your instructions are probably a mess. Mine were too. XML fixed it.

Most AI agents break because their instructions are too long, too vague, or too human. I used to write like that too — until I switched to XML and finally got the control I needed.