How Todoist is Built: Inside Doist's Bootstrapped, Async-First Engineering Culture

How Todoist is Built — with Gonçalo Silva

TLDR: Doist, the company behind Todoist, operates with 100 employees across 40 countries, no office, and no outside funding, yet serves 25 million users and competes with billion-dollar funded competitors. Their CTO reveals how async-first communication and functional team structures make this possible.

This interview with Gonçalo Silva, CTO of Doist, is one of the most interesting looks at alternative ways to build a software company. While the industry obsesses over raising the next round and scaling headcount, Doist has quietly built Todoist into a tool used by 25 million people while staying bootstrapped, profitable, and small.

Let's start with the numbers that should make you think differently about scaling. One hundred employees. Forty countries. No office. Profitable. Competing against tools backed by hundreds of millions in funding. The conventional wisdom says this shouldn't work. You need proximity for collaboration. You need funding for growth. You need headcount for features.

The secret sauce isn't a secret at all: it's async-first communication. Not "async-friendly" or "async-optional," but async as the default. When your team spans time zones from New Zealand to California, synchronous communication becomes a constraint rather than a tool. Doist built their own product, Twist, specifically to enable this workflow. Decisions are made in written form. Context is preserved in threads. Knowledge doesn't disappear when a meeting ends.

The functional teams versus squad-based execution discussion is particularly relevant for engineering leaders. Many companies have adopted the Spotify model of cross-functional squads, but Doist takes a different approach. They maintain functional expertise while organizing work around execution. This preserves deep domain knowledge while still enabling cross-team collaboration on features.

What's refreshing is the honest take on balancing opinionated product vision with user feedback. Todoist has strong opinions about how task management should work. They're not building everything users request. This is countercultural in an era of feature flag everything and let data decide. But it's also how you build a product with a coherent point of view rather than a collection of disconnected features.

The discussion on AI's impact on engineering work is timely. Rather than treating AI as either savior or threat, Silva takes a pragmatic view of how it's changing day-to-day development work at Doist. It's a tool that amplifies capability, not a replacement for judgment.

For architects and engineering leaders, the Doist model challenges assumptions about what's necessary for building successful software. You don't need co-location. You don't need constant meetings. You don't need venture funding. You need clear communication, strong opinions about your product, and the discipline to work asynchronously.

Key takeaways:

• Async-first communication enables effective collaboration across extreme time zone distribution
• Bootstrapped companies can compete with heavily funded competitors through focus and efficiency
• Functional team structures can work alongside cross-functional execution
• Strong product opinions create coherent user experiences, even if it means saying no to requests

Tradeoffs:

• Async-first enables global talent access but requires exceptional written communication skills from everyone
• Staying bootstrapped preserves control and forces profitability but limits growth speed
• Opinionated product design creates coherent UX but may alienate users wanting different workflows

Link: How Todoist is Built — with Gonçalo Silva

This article was generated from a Substack newsletter. The summary is based on the original content and includes editorial commentary and analysis.