Jack Dorsey's Block Claims AI Tool Handles 15% of Coding Tasks
Block's new AI tool Builderbot is already doing 15% of the company's code work, signaling a big shift in how tech teams operate.
Jack Dorsey's payments and tech company Block is making waves in the AI coding space with a homegrown tool called Builderbot. According to Block's head of AI capabilities, Brad Axen, the tool is already responsible for handling roughly 15% of the company's coding workload — which is a surprisingly significant chunk for any large engineering organization.
Axen described Builderbot as the "missing layer between AI coding tools and how engineering actually works at scale." That's a pretty telling phrase. Most off-the-shelf AI coding assistants are great for individual developers writing snippets or debugging on their own, but they tend to fall apart when you need them to function inside the messier, more complex reality of a big company's engineering pipeline. Builderbot is Block's attempt to bridge that gap.
Read more FG Communities Expands in Western North Carolina With New Acquisition →
For everyday readers, think of it this way: AI tools like GitHub Copilot help individual coders write faster, but they don't necessarily understand your company's internal systems, workflows, or deployment processes. What Block appears to be building is something more deeply integrated — a tool that actually understands how engineering gets done at Block specifically, not just coding in general.
This development matters beyond Block's walls. Companies across the tech industry are racing to figure out how AI can genuinely replace or augment software development at scale, not just assist individual engineers. If Block's 15% figure holds up and grows, it could pressure other firms to accelerate their own internal AI tooling efforts. It also raises real questions about how engineering team sizes might change as these tools mature.
Whether Builderbot stays an internal tool or eventually becomes a product Block sells to others remains to be seen — but either way, it's a signal that enterprise-grade AI coding infrastructure is moving fast. Continue reading at Cointelegraph.
