Supercharge your "vibe coding" with a structured Github Template
For - Technical founders, Developers, Engineering Managers.
We move fast.
On our team, it’s normal to code a feature in the morning and push it to production by night. That speed gives us an edge. But it also comes with hidden costs.
A few weeks ago, one of our junior developers shipped a feature that worked flawlessly in our dev environment. But when our product manager tested it in production, it broke backward compatibility. Everything looked fine locally. The code was clean. But in our rush to ship, we skipped the pause—that crucial moment to think about edge cases, system impacts, and downstream effects.
It wasn’t just a technical miss; it was a sign that even with the best AI tools, our brains needed a structured prompt to think beyond the immediate code. And that’s where a simple GitHub template transformed how we build.
TL;DR (Key Takeaway)
The Incident That Changed Everything
I’ll never forget the look on our product manager, Shavin’s face when he realized that a feature we’d shipped that morning had just broken backward compatibility in production. He wasn’t angry- he was disappointed.
We’d promised this feature to a key customer, and everything worked perfectly on our staging environment. But in production, it failed in ways none of us had predicted.
It wasn’t anyone’s fault. The code was clean, the logic sound. But in our rush to ship, fueled by the speed of modern AI tools, we’d skipped the pause that would have forced us to think about system-level impacts. The second and third-order consequences that come from integrating a feature into a live product with real users.
We were coding faster than ever, but we weren’t thinking deeper.
The Pause That Made Us Faster
That’s when Gayanga, one of our teammates, suggested a simple but powerful idea: a GitHub Issue Template. At first, it felt like bureaucracy. Another step. But he reframed it: “It’s not overhead,” he said. “It’s a prompt - for our brains. A way to slow down just enough to think about the bigger picture.”
He was right. AI tools like Cursor and Windsurf can generate code at lightning speed, but they can’t prompt us to think about what might break in production, or how this change might affect other parts of the system. They excel at local optimizations but struggle with system-level reasoning.
Your Brain as a Human LLM
Our brains hold years of context: every messy migration, every late-night bug fix, every lesson learned from production outages. But like an AI model, that knowledge needs a prompt to surface. Without it, we’re likely to jump from idea to code without considering the consequences.
Recommended by LinkedIn
Our Github Feature Template
Here's what it looks like:
Feature Description
What is this feature? Drop a short description here. Even if there is already existing product docs, the developer write down from their own words what the feature is.
Solution Plan
In point form:
Quality and Speed
We answer:
What it Changed for Us
Since implementing this template, we've seen concrete improvements:
It's not just about writing better features - it's about building a better thinking process around feature development. A habit of intentional engineering.
Counter-intuitively, adding this structured pause actually made us ship faster overall. The time we invest upfront pays dividends by eliminating the costly back-and-forth of fixing issues post-deployment.
And for a team that ships fast, that pause can make all the difference.
Why “RideIsIt”?
The name of this newsletter is inspired by a quote from Terry, a surf instructor in Sophie Kinsella’s book Burn Out. Terry reminds his students that it’s not just about catching the perfect wave - it’s about enjoying the ride, every time.
That philosophy resonates with me as a founder: building a startup is unpredictable, sometimes messy, but the ride itself is where the real growth happens.
What’s one challenge you’re navigating right now? Hit reply—I’d love to hear your story and how you’re riding the waves of building and growth.
