Luís Rodrigues Advocates for AI-Assisted PRD Creation to Streamline Product Development

In a recent LinkedIn post, Luís Rodrigues discusses the inefficiencies of traditional product development processes, particularly the manual creation of Product Requirements Documents (PRDs) after discovery calls. Rodrigues argues that modern tools, when used strategically, can significantly reduce the burden on Product Managers.

Rodrigues begins by highlighting the dated nature of manual summary writing and PRD creation, stating:

Writing meeting summaries from scratch in 2026 is embarrassing. Writing a PRD from scratch after a good discovery call is worse.

He elaborates on the time-consuming nature of this task, explaining his personal shift away from the practice. Rodrigues shares his experience using a tool called Granola, which records and transcribes calls, while still allowing him to take his own notes during the conversation.

Streamlining PRD Generation with AI Assistance

The core of Rodrigues’s argument centers on leveraging AI, specifically mentioning Claude, to assist in drafting PRDs. He outlines a workflow where Granola, integrated with Claude, utilizes the call transcript and his own notes to generate the document. This approach aims to bypass the need for extensive manual re-writing and context re-explanation.

According to Rodrigues, the process involves:

1. Pulling the discovery call transcript from Granola.
2. Instructing Claude to draft a PRD, including sections for problem, user, requirements, open questions, and out of scope.
3. Crucially, quoting the specific transcript line behind every requirement.

Rodrigues emphasizes a key detail that he believes prevents AI hallucinations:

If a requirement has no line behind it, leave it out. The last two lines are the real trick to avoid hallucinations.

This method, as Rodrigues points out, eliminates the need for copy-pasting and re-explaining context to an AI that wasn’t present during the live conversation. The AI drafts the PRD, but the Product Manager retains final decision-making authority.

The Role of the Product Manager in the AI Era

Rodrigues is clear that AI serves as an assistant, not a replacement, for human judgment. He states:

Claude drafts it. I decide what ships.

This highlights his perspective that AI tools can accelerate the initial drafting phase, allowing Product Managers to focus on refining and validating the requirements rather than spending hours on initial documentation. He concludes by posing a question to his network:

“PMs: what’s the worst PRD you’ve ever had to write from a call?”

Rodrigues encourages others to repost his insights to help reduce the toil associated with meeting follow-ups and documentation, framing it as a way to help people “suffer less in meetings.” His post advocates for a more efficient, AI-augmented approach to product development documentation.