In a recent LinkedIn post, 🌀 Luca Rossi shares insights from a conversation with the founder and CEO of Unblocked, delving into the evolving landscape of software development and the impact of artificial intelligence.
The discussion, featured on the Refactoring podcast, touched upon several critical areas where current trends in AI are not necessarily yielding expected productivity gains. Rossi highlights that the sheer volume of tokens processed by AI models doesn’t automatically equate to enhanced output.
As 🌀 Luca Rossi notes:
I had a conversation with the founder and CEO of Unblocked about the latest trends in software development.
The conversation zeroed in on key questions facing development teams today. Rossi emphasizes the group’s exploration into how AI can be leveraged to ensure code quality, a persistent challenge in fast-paced development cycles. This includes understanding the practical application of AI tools not just in theory, but in real-world production environments.
The Productivity Paradox: More Tokens, Not More Output
One of the central themes explored in Rossi’s shared insights is the notion that increased AI processing power, measured in tokens, does not directly correlate with a proportional rise in developer productivity. This suggests a more nuanced approach is needed to integrate AI effectively into development workflows. Rossi points out that the focus may need to shift from raw processing capacity to how AI is guided and applied.
Ensuring Code Quality with AI
According to 🌀 Luca Rossi, a significant portion of the discussion revolved around the practicalities of guaranteeing code quality when integrating AI. This involves not just using AI for code generation, but also for review and validation processes. The challenge lies in ensuring that AI-assisted code meets rigorous standards before being deployed to production.
We discussed: • Why more tokens are not translating into more productivity • How to guarantee code quality with AI and • What’s the role of context in making your agents succeed
🌀 Luca Rossi further elaborates that the effectiveness of AI agents is heavily dependent on the context provided to them. This underscores the importance of well-defined prompts and sufficient background information for AI tools to function optimally.
Real-World Applications and Use Cases
The conversation also delved into how companies like Unblocked are implementing these AI strategies. Rossi shared that they explored the team’s setup, their approach to navigating these new technologies, and specific use cases detailing how features are being shipped into production using AI. This practical perspective offers valuable lessons for other organizations looking to adopt similar technologies.
We also explored how the Unblocked team is doing it, navigating their setup and a few use cases about how their shipping features in production.
The full discussion provides a comprehensive look at the current state and future potential of AI in software development, as articulated by 🌀 Luca Rossi and his podcast guest.
📝 About This Content
This article is based on insights shared by 🌀 Luca Rossi on LinkedIn.
📅 Originally posted on September 8, 2026 | View original post on LinkedIn →