.svg)
At this year's Google I/O conference, Google dropped two tools that have developers talking. Stitch and Jules aren't just another pair of AI assistants—they're tackling real problems that anyone who's spent late nights debugging code or wrestling with UI layouts will recognize.
Remember the days when creating a decent-looking interface meant hours of tweaking CSS margins and hunting down the perfect color palette? Stitch changes that equation entirely.
This AI-powered design tool lets you describe what you want—or even upload a rough sketch—and generates clean HTML and CSS code. But here's what makes it interesting: you can switch between Google's Gemini 2.5 Pro and Flash models depending on how complex your project is.
The workflow is surprisingly straightforward. You tell Stitch what kind of interface you need, maybe throw in some specifics about colors or layout preferences, and it spits out working code. Need a goal-tracking app interface? how to draw stitch drawing? Done. Want to experiment with a new navigation design? No problem.
The Figma integration is where things get practical. Instead of bouncing between tools, your team can collaborate on designs right where they already work. And when you're ready to code, you can export everything to your preferred editor.
What's refreshing is that Stitch doesn't assume you're a design expert. Whether you're a backend developer who needs to mock up a quick frontend or a designer who wants to see their ideas in code, the tool adapts to your skill level.
Jules: The Debugging Detective
While Stitch handles the visual side, Jules dives into the messy reality of maintaining code. Anyone who's inherited a legacy codebase or spent hours tracking down a mysterious bug will appreciate what Jules brings to the table.
Jules doesn't just find bugs—it understands context. During Google's demo, they showed Jules upgrading a website from Node.js 16 to version 22. This wasn't just a simple find-and-replace operation. Jules:
The whole process happened without human intervention, but with human oversight at every critical step.
The tool connects directly to GitHub, so it understands your existing project structure. When you point Jules at a repository, it clones everything to a cloud VM and starts analyzing. The AI (currently Gemini 2.5 Pro) looks for patterns, identifies potential issues, and suggests improvements.
But here's the key part: Jules doesn't just make changes and hope for the best. It generates detailed diffs showing exactly what it wants to modify. You review, approve, and only then does Jules create pull requests and handle the merge process.
One feature that caught my attention is the audio summaries. After Jules completes work, it generates a spoken explanation of what changed and why. For teams where not everyone has time to read through commit logs, this could be genuinely useful.
The software industry has been talking about AI-assisted coding for years, but most solutions feel like fancy autocomplete. Stitch and Jules address different pain points: the tedium of repetitive design work and the complexity of maintaining large codebases.
Neither tool claims to replace developers. Instead, they handle the tasks that eat up time without adding much value, like converting a design concept into basic HTML or tracking down compatibility issues during framework upgrades.
Google isn't alone in this space. GitHub Copilot has become commonplace for many developers, and tools like Cursor are pushing the boundaries of AI-assisted coding. What sets Stitch and Jules apart is their focus on complete workflows rather than just code generation.
Stitch doesn't just generate CSS—it integrates with design tools. Jules doesn't just suggest fixes—it handles the entire upgrade process from planning to deployment.
Both tools are still evolving. Google mentioned plans to let users switch between different AI models for various tasks, which could make them more flexible as different models prove better for specific use cases.
The real test will be how these tools perform in messy, real-world scenarios. Can Stitch handle complex design systems? Will Jules manage upgrades for codebases with unusual architectures or custom frameworks?
Early access to Jules is available now, and Stitch should follow soon. For developers tired of spending time on routine tasks, these tools might be worth the experiment.
The future of software development isn't about AI replacing programmers—it's about AI handling the repetitive work so humans can focus on solving interesting problems. Stitch and Jules feel like solid steps in that direction.
.svg)
