Coinbase CEO Brian Armstrong Says AI Now Writes 40% of Exchange’s Code, Targeting 50% by October

Coinbase Leans on AI for Code Development

Coinbase CEO Brian Armstrong revealed that artificial intelligence now writes over 40% of the code powering the crypto exchange’s systems — more than double the share in April.

Armstrong added that while AI-generated code must still be reviewed and understood by engineers, he believes the company should use AI “as much as we possibly can” in a responsible way.

His next milestone: reaching 50% AI-generated code by October.

The update comes just weeks after Coinbase emphasized its mission to become an “AI-Native” company, signaling a shift toward workforce augmentation rather than wholesale replacement of employees.

Related: Coinbase CEO Brian Armstrong Calls for Overhaul of Token Listing Process Amid Surge in New Tokens

Why Coinbase Is Betting Big on AI

Armstrong has long argued that AI can boost productivity rather than eliminate jobs. Engineers at Coinbase now use coding assistants like Copilot, Claude Code, and Cursor to accelerate development.

According to Armstrong:

• Single engineers can now refactor or build new codebases in days instead of months.

• AI enables efficiency gains that were “not possible 12 months ago.”

On John Collison’s Cheeky Pint podcast, Armstrong admitted he let go of engineers who resisted using AI, after mandating in Slack that the company adopt it across workflows.

Change in the percentage of Coinbase’s code that is AI-generated. Source: Brian Armstrong

Coinbase Hiring Surge in AI-Focused Roles

Despite integrating AI deeply into operations, Coinbase is actively hiring. Of its roughly 350 open positions, nearly half are for engineering and development roles.

• 93 openings are for back-end engineers, many referencing AI expertise.

• Customer experience is the largest non-tech hiring category, with 56 roles available.

This reflects Armstrong’s view that AI is not eliminating jobs, but rather reshaping the skillsets Coinbase values most.

Industry Impact: AI vs. Crypto Talent

While large-scale layoffs in crypto have slowed, recruiters say crypto hiring has tightened since 2022. Many attribute this to capital and talent flowing toward AI startups.

“Developers and entrepreneurs follow the money and excitement, and right now AI is soaking up both,” CryptoJobsList founder Raman Shalupau told bitcoinworld.news.

Still, analysts argue that fears of mass AI-driven unemployment are overblown. PwC researchers and U.S. AI policy chief David Sacks suggest AI will more likely enhance productivity rather than cause systemic job losses.

Conclusion: Coinbase Is Building an AI-Native Workforce

With 40% of code already AI-written and hiring focused on AI-capable engineers, Coinbase is positioning itself at the intersection of crypto and artificial intelligence.

If Armstrong’s 50% goal is achieved by October, Coinbase may emerge as a case study for how AI integration can accelerate innovation in highly regulated, fast-moving industries like crypto.\

Related: Crypto Market Recap: Coinbase Makes TIME 100 List, Buterin Unveils New ID Tech, XRP Rallies on SEC Update

FAQs

1. How much of Coinbase’s code is written by AI?

Brian Armstrong revealed that AI now writes over 40% of Coinbase’s code, with a goal of reaching 50% by October.

2. Does Coinbase still use human engineers?

Yes. While AI generates code, human engineers review, test, and refine it to ensure reliability and security.

3. Why is Coinbase integrating AI into its development process?

Coinbase aims to boost efficiency, with single engineers now able to refactor or build codebases in days instead of months using AI tools.

4. Is Coinbase reducing jobs because of AI?

No. Coinbase is actively hiring — nearly half of its ~350 openings are for engineers, many requiring AI expertise.

5. What tools does Coinbase use for AI-assisted coding?

Engineers at Coinbase use AI tools like Copilot, Claude Code, and Cursor to accelerate code development.