Daniel Doubrovkine, who runs Augmented Engineering at Shopify, in a recent episode of The Hangar DX podcast, explained:

• How they use AI tools to transform developer experience
• How they’re betting on AI as a productivity multiplier in the engineering organization, and why
• How managers and ICs alike can adapt

What is Augmented Engineering?

Daniel describes his team as part of Shopify’s Developer Productivity organization, but with a clear AI-first focus. Augmented Engineering, he says, means either augmenting engineers with AI tools or augmenting dev tools with AI.

The distinctive aspect of our work is that we’re specifically focused on augmenting engineers with AI wherever it makes sense so they can be more productive.

Reflexive Use of AI

Many companies talk about adopting tools like Claude, Copilot, and Cursor. At Shopify, Daniel says, that’s the default, what they call the reflexive use of AI. Instead of thinking “maybe I can use AI here,” engineers at Shopify try an AI tool first for any problem.

We think of AI as a companion or an agent that’s with you at every step. That’s what we’re building: tooling that lets AI help everywhere, from writing code to reviewing it to maintaining it.

AI-Driven Developer Productivity

Shopify’s mission is to make commerce better for everyone, and the Developer Productivity organization’s mission is to “make commerce better for everyone.”

They’re not trying to find problems for AI—they’re tackling genuine, large-scale pain points that are hard to solve only with humans. For example, Shopify runs hundreds of thousands of tests, but test coverage is uneven. One goal is to use AI to generate high-quality, missing tests automatically. This is crucial groundwork for a future where AI agents can safely refactor production code.

Another classic challenge is flaky tests. Even with robust manual efforts—like isolating flaky tests and assigning them to developers—flakiness remains a constant drag on productivity. Daniel’s team is exploring whether AI can help finally put an end to that persistent problem.

AI Tools in Action: Shopify’s Roast

One of Shopify’s initiatives is Roast, an AI workflow tool they have open-sourced. They use it internally to automate test grading and test generation. The workflows even feed into each other: if a test gets a failing grade, another AI workflow can generate an improved version, iterating until it passes with high marks.

The pace of improvement in these tools is remarkable. That said, there’s no illusion about AI being a silver bullet.

The AI is definitely getting better fast. I don’t know where it ends, but I certainly want to be there.

We still need experts to recognize that it did a good job, though. It might produce something extraordinary with security problems, and we still need to be experts to recognize that. 

Advice for Companies on Adopting AI

Daniel emphasizes that adopting AI isn’t just about buying the latest tools—it requires a real shift in mindset. But companies also need to make AI tools accessible and integral to workflows.

He also suggests changing their hiring process:

Ask candidates how they’ve used AI in their workflow, and hire those who have tried the tools, persevered, and created amazing results. Those are the people you want! 

Learn, share, and collaborate on topics related to developer productivity. Join Hangar DX, a community of senior DevOps and senior software engineers focused on developer experience.