---
title: Founders Should Still Write Code
description: "As a technical founder, I write code every day. It keeps me honest about complexity, connected to the product, and dangerous in the AI era where code output matters."
canonical: https://nowah.xyz/blog/founders-should-still-write-code
lastModified: "2026-08-07T08:32:23.286Z"
---

# Founders Should Still Write Code

As a technical founder, I write code every day. It keeps me honest about complexity, connected to the product, and dangerous in the AI era where code output matters.

The conventional wisdom is clear: at some point, founders should stop coding and start "leading." Delegate the technical work. Focus on strategy, fundraising, and hiring. Get out of the codebase.

I think this advice is wrong, especially now.

## Why founder coding builds better products

![Illustration for this section](https://pics.nowah.xyz/website-media/founder-050-img-1.webp)

When you write code daily, you feel the complexity of your own product. Not abstractly, through architecture diagrams and sprint reviews. Directly, through the experience of implementing features, fixing bugs, and working within the system you built.

This direct contact changes how you make product decisions. You cannot handwave away a three-week engineering estimate when you understand exactly what those three weeks involve. You cannot ask for "a simple change" when you know there is no such thing in the area of the codebase that needs changing.

Founders who stop coding gradually lose calibration. They start making decisions based on outdated understanding of the system's complexity. They approve timelines that are unrealistic because they no longer feel the work.

## The danger of losing context

Every day you do not touch the code, your mental model of the system drifts further from reality. Abstractions accumulate. Assumptions go unchecked. The gap between your understanding and the actual state of the codebase widens.

This gap is dangerous because it degrades decision quality. The best product decisions come from understanding the codebase firsthand. What is technically easy? What is surprisingly hard? Where are the fragile points? What would break if we changed this?

These questions have obvious answers when you are in the code. They have speculative answers when you are not.

## When to code and when to lead

![Supporting diagram](https://pics.nowah.xyz/website-media/founder-050-img-2.webp)

I am not arguing that founders should spend all their time coding. Strategy, hiring, user conversations, and business development are essential and cannot be neglected.

The sweet spot is maintaining enough coding involvement to stay calibrated without crowding out the founder-specific responsibilities that nobody else can do. For me, that means writing code every day, but not all day. Core architecture decisions, complex features, and critical fixes. Not routine tasks that the team handles better.

The key is that the coding is not optional busy work. It is a strategic input to better leadership.

## The AI era multiplier

The argument for founder coding is stronger now than ever. AI coding assistants multiply the output of every engineer, founders included. The leverage per hour of founder coding has increased dramatically.

A founder who spends two hours coding with AI assistance can produce meaningful features, architectural improvements, or bug fixes that would have taken a full day three years ago. The opportunity cost of those two hours is lower, and the output is higher.

In a small team, founder coding is not just philosophically valuable. It is structurally necessary. The team is small enough that every person's contribution matters, including the founder's.

## The respect factor

There is a practical benefit that rarely gets discussed: team respect. Engineers respect leaders who understand their work at a technical level. Not because coding is impressive, but because it demonstrates that you respect the complexity of what the team does.

A founder who says "this should be easy" without understanding the code loses credibility. A founder who says "I looked at the code and I understand why this is complex, here is how we should prioritize it" earns respect and makes better decisions.

If you are a technical founder, keep shipping. Your team, your product, and your judgment will be better for it.

---

Nowah is an AI travel agent that searches and books real flights and hotels through conversation — no filters, no thirty open tabs. [Plan your next trip](https://app.nowah.xyz).