Don't Panic Labs

10/14/2026

“Are you in love with the problem or are you in love with the solution?” Lindsey Hruby-Yardley asks every founder she meets that question.

On LFG with AQI, Lindsey talks about the work that should happen before a founder builds anything: finding evidence that people want the idea and will pay for it. She also covers her nearly 22 years in the Nebraska Air National Guard, and the grants and funding sources Nebraska founders often don’t know exist.

Watch the full episode: https://www.youtube.com/watch?v=1X1_Oeb-Zh4

10/08/2026

Last night we were saddened to hear about the passing of Margaret Hamilton, who led the team that wrote the on-board flight software for Apollo.

In 1965 she became the first programmer hired for the Apollo project at MIT. At that time, no school taught software development, so she and her team worked it out as they went. She also pushed for the people writing software to be treated as engineers. Her colleagues teased her about the phrase “software engineering” for years, until the rest of the industry caught up with her.

Three minutes before Apollo 11 touched down, the guidance computer was overloaded and alarms went off. Her team had built the software for that moment. It dropped lower-priority tasks, kept the ones needed for landing, and gave the crew and Mission Control what they needed to make the call to keep going.

Software engineers now work in a discipline she fought to legitimize. Her example of designing for the failure you hope never comes still ripples through our industry.

Thank you, Margaret.

09/15/2026

In her first blog post, Bianca Zongrone Jefferson walks through building a synthetic user for a system redesign project.

When a designer works alone, it is easy to lose sight of the user's needs. Building a synthetic user helps maintain a clear user perspective.

09/09/2026

GitFlow starts fights in team chats because everyone means something different by it.

Chad Michel breaks down the two-branch version we run on our own projects: develop for QA, staging, and preview, and main for production only.

Chad Michel walks through a two-branch GitFlow variant, develop and main, that keeps deploys predictable for small and medium teams.

08/25/2026

Bill made an executive call on Friday: red is the new blue at Don't Panic Labs.

08/18/2026

Chad Michel got OpenAI's Codex CLI running locally through Ollama. This means no cloud API and no code leaving his machine. Check out how he did it.

Chad Michel shows how to run OpenAI's Codex CLI locally with Ollama, from installing the gpt-oss model to setting up a persistent local profile.

08/04/2026

Read how Chad Michel treats the TanStack Query cache as the source of truth for the UI and lets the server catch up whenever it can.

Chad Michel builds an offline-first flow with TanStack Query, treating the local cache as the source of truth so the UI stays fast even when the network drops.

07/29/2026

You can tell a kitchen is messy without being a chef. The same goes for code.

Chad Michel wrote a field guide that walks through thirteen smells you can catch without knowing a programming language.

A field guide to spotting bad code without writing any. Thirteen smells anyone can catch.

07/22/2026

For Chad, VMs felt like overkill and Kubernetes was too much infrastructure for the project. But an Azure App Service with containers hit the sweet spot.

Check out his post as he shares how he deployed a single container bundling the Angular frontend, NestJS API, and Caddy, all built in GitHub Actions and pushed to Azure Container Registry.

Chad Michel walks through running a single Docker container on Azure App Service: Angular, NestJS, and Caddy, built in GitHub Actions with zero-downtime swaps.

07/16/2026

Bill Udell has been watching organizations reach for a Responsible Use Policy before they have decided what they actually believe about AI.

In his latest post, he makes the case that governance comes second. Leadership has to decide who the organization intends to become first.

Most organizations start their AI work by writing a policy. Bill Udell argues that governance comes second. Leadership has to decide who you intend to become first.