TL;DR
A Jane Street designer now relies more on Claude AI than Figma for creating prototypes and features. This change enhances productivity and accelerates development, but raises questions about review and creativity.
A designer at Jane Street has increasingly replaced Figma with Claude AI for creating prototypes and implementing features, citing significant workflow improvements and faster iteration cycles.
The designer reports that over the past two months, their reliance on Claude AI for prototyping and development has grown, surpassing traditional tools like Figma. They now use Claude to generate functional prototypes directly in code, enabling rapid iteration without the need for extensive mockups or documentation. For example, a recent prototype added LLM prompting to a JSQL input, which Claude refined through unlimited iterations, saving days or weeks of engineering work. Initially, AI was used for smaller tasks, but improved models and experience have expanded its role to larger, more complex projects, including prototypes with over 2000 lines of code and new app designs. This workflow empowers engineers and designers alike, facilitating quick validation of ideas and reducing reliance on mockups and static documentation. However, the approach introduces new challenges, particularly around review processes, as fully baked features are presented for feedback, potentially limiting collaborative input during development. The designer emphasizes the importance of treating prototypes as living proposals, with review focused on user experience and design rather than code correctness.
Why It Matters
This shift signifies a potential transformation in design and development workflows at tech companies, where AI tools like Claude can substantially reduce time and effort in prototyping and implementation. It may influence how teams evaluate ideas, prioritize rapid iteration, and collaborate across disciplines. However, it also raises questions about the balance between automation and creativity, as well as review processes that adapt to AI-generated prototypes.
AI Fashion Design & Prototyping Unleashed: The Quick-Start Guide for Rapid Design & Iteration: Master CLO 3D, Browzwear, Midjourney & Digital AI Tools … & Faster Workflows (Fashion Metaverse)
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Background
Historically, designers relied heavily on tools like Figma and detailed documentation, especially for larger projects. The rise of large language models (LLMs) and AI-driven development has begun to challenge these norms. At Jane Street, a quantitative trading firm known for its technical rigor, a designer’s recent experience highlights how AI is reshaping workflows. The transition began with small tasks but has accelerated recently due to improvements in AI models and increased familiarity with the tools. This mirrors broader industry trends where AI is increasingly integrated into software development and design processes.
“Using Claude to make these ideas real I’m making it a lot easier for others to evaluate them—they can just use it.”
— Jane Street Designer
“Prototypes are living proposal docs, the code is disposable, and a reviewer’s job is to give feedback about the design and user experience.”
— Jane Street Designer
AI Engineering: Building Applications with Foundation Models
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What Remains Unclear
It is still unclear how widespread this approach will become across teams and organizations, and how review processes will evolve to accommodate AI-generated prototypes. There are also questions about long-term impacts on creativity and collaboration, which remain to be explored.
Conversational Multi-Agent Systems: Collaborative AI Agents for Developers and Testers (Context Engineering for Multi AI systems Guide Book 1)
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
What’s Next
Next steps include refining review practices to balance feedback on design versus implementation, exploring broader adoption of AI-driven prototyping, and assessing impacts on team dynamics and project outcomes. Further developments in AI models will likely expand capabilities and influence workflows.
Low-Code AI: A Practical Project-Driven Introduction to Machine Learning
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Questions
How reliable are AI prototypes compared to traditional mockups?
AI prototypes can be highly functional and fast to produce, but they may require review and refinement to ensure they meet user needs and quality standards. They are tools for rapid validation rather than final designs.
Will this approach replace Figma entirely?
Not necessarily. While AI is replacing some aspects of design and prototyping, tools like Figma still offer valuable visual and collaborative features. The shift is toward integrating AI to accelerate workflows, not eliminating existing tools altogether.
What are the risks of relying heavily on AI for design and development?
Potential risks include reduced creative exploration, over-reliance on AI outputs, and challenges in collaboration and review processes. Ensuring human oversight and iterative review remains essential.
How might this change team collaboration?
Teams may need to adapt review workflows to focus more on design intent and user experience rather than code correctness. Clear communication about prototypes as living documents will be crucial.
Source: Hacker News
