The non-technical PM’s guide to building with Cursor | Zevi Arnovitz (Meta) - Lenny's Podcast Recap
Podcast: Lenny's Podcast
Published: 2026-01-18
Duration: 1 hr 15 min
Guests: Zevi Arnovitz
Summary
Zevi Arnovitz, a product manager at Meta, shares how non-technical individuals can build products using AI tools like Cursor. He details his workflow and techniques that enable him to develop and ship real products despite having no technical background.
What Happened
Zevi Arnovitz, a product manager at Meta without a technical background, has mastered the art of using AI tools to build and ship products. He utilizes tools like Cursor and Claude code to enable non-technical people to create sophisticated products, effectively collapsing traditional job roles. His AI workflow involves the use of multiple AI models for different tasks, such as Claude for planning and Gemini for UI design, making the process accessible and efficient for non-technical individuals.
One of the key techniques Zevi employs is using slash commands to automate prompts within his development workflow. This allows him to efficiently manage tasks like creating issues, exploring ideas, planning, executing, and updating documentation. Zevi also uses a peer review process where different AI models review each other's code, similar to team leads in a company, which helps in catching mistakes and refining the product.
Zevi emphasizes that this is an excellent time to be a junior in tech, as AI tools enable individuals to independently build startups. He shares his experience of using AI to prepare for his Meta PM interviews, showcasing how AI can be leveraged for personal growth and career advancement.
In his workflow, Zevi creates a CTO-like AI agent to challenge him on technical decisions, ensuring a robust development process. He uses Cursor's slash commands to streamline his workflow, allowing him to quickly execute tasks and stay in the flow of development. This system has enabled him to develop projects like StudyMate, a platform for students to upload study materials and create interactive tests.
Zevi's approach to AI tool usage highlights the importance of starting slow and gradually becoming more comfortable with code. He notes that while tools like Bolt and Lovable simplify the development process, they limit user control, whereas Cursor and Claude offer more comprehensive decision-making capabilities.
The episode also touches on how AI has impacted the PM role, with AI tools enhancing productivity and learning rather than replacing human skills. Zevi suggests that PMs can play a larger role in the future, leveraging AI to build and ship products without deep technical expertise.
Key Insights
- AI tools like Cursor and Claude enable non-technical individuals to build and ship sophisticated products by automating tasks such as planning, executing, and updating documentation through slash commands.
- Zevi Arnovitz uses a peer review process where different AI models review each other's code to catch mistakes and refine products, mimicking the role of team leads in a company.
- Creating a CTO-like AI agent to challenge technical decisions is part of Zevi Arnovitz's workflow, ensuring a robust development process for projects like StudyMate, a platform for interactive student tests.
- AI tools have transformed the product manager role by enhancing productivity and learning, allowing PMs without deep technical expertise to build and ship products independently.