Lawyer,Cardiologist and Road Tech take pole positions at a Hackerthon.
Photo by BoliviaInteligente on Unsplash
Tech March 25, 2026 · By Garv

Lawyer,Cardiologist and Road Tech take pole positions at a Hackerthon.

Interesting backgrounds of the winners of Anthropic's "Built with Opus 4.6" Claude Code Hackathon and what it means for the future of innovation in tech.

Share: Twitter Facebook LinkedIn

While doom-scrolling on Reddit recently, a post about a lawyer winning a hackathon caught my eye—and I had to dig into the details of this unlikely outcome.

Anthropic, the company behind Claude, hosted a virtual hackathon where they selected 500 participants from over 13,000 applicants. Each participant was given $500 worth of Claude API credits and tasked with building something “meaningful” in just six days.

As the name of the hackathon suggests, the competition was built around Opus 4.6—Anthropic’s most powerful model—featuring a 1 million-token context window and enhanced agentic coding capabilities. Submissions were evaluated by six judges from the Claude team (their names are immaterial here) based on technical innovation, implementation quality, and potential impact.

Of the five winners—each receiving $100,000 in Claude API credits—only one was a software engineer by trade. The top three included a lawyer from California who built a permit processing app, a cardiologist from Brussels who developed a patient follow-up tool, and a road technician from Uganda who created an infrastructure assessment system.

The overarching point is this: domain expertise proved more valuable than traditional knowledge of software development methodologies or coding ability.

My take? AI has the power to enable people who understand real-world problems to build solutions—without needing deep programming expertise.

That said, two things can be true at once. As much as “vibe coding” may be the current zeitgeist, there is still a clear need for experienced software professionals—people who can bridge domain knowledge and technology to create secure, scalable, and meaningful systems.

Controls, process, and accountability still require humans in the loop—especially for critical systems.

Not because humans are perfect (far from it), but because when things inevitably go wrong, there needs to be accountable people—not just vague references to “AI hallucinations.”

How did you find this article?

Garv

Garv

Blog administrator and primary author

An unhandled error has occurred. Reload 🗙

Rejoining the server...

Rejoin failed... trying again in seconds.

Failed to rejoin.
Please retry or reload the page.

The session has been paused by the server.

Failed to resume the session.
Please retry or reload the page.