The NLX (Natural Language Experience) Design Stack
by Aparna Chennapragada • Chief Product Officer, AI Product Strategy at Microsoft
Former CPO at Robinhood, VP at Google (Lens, AR, Assistant), Engineering Leader at Akamai, Board Member at eBay and Capital One.
🎙️ Episode Context
Aparna Chennapragada, CPO at Microsoft, discusses the shift in product development driven by AI. She explores the transition from graphical interfaces to natural language experiences (NLX), the necessity of prototyping as a primary specification tool, and frameworks for evaluating the timing and viability of zero-to-one products. The conversation covers the balance between innovation and governance in enterprise AI, the evolving role of product managers, and the concept of collaborative intelligence between humans and agents.
Problem It Solves
Prevents 'Frankenstein products' where AI is just a raw model attached to an app without design structure, ensuring usability and trust.
Framework Overview
A framework defining the new principles and UI constructs required for Natural Language Interfaces, moving beyond simple chat to structured, designed conversations.
🧠 Framework Structure
Prompt as UI: The prompt is a distinc...
Editable Plans: Agents should return ...
Show the Work: Visualize the agent's ...
Designed Follow-ups: Proactively sugg...
When to Use
When designing AI agents, chatbots, or conversational interfaces to ensure they are intuitive and effective.
Common Mistakes
Assuming 'the model is the product' and neglecting design; making the AI a black box without feedback; failing to guide the user after an output.
Real World Example
Designing a research agent that doesn't just answer but proposes a plan for how it will gather information, or an image generator suggesting logical stylistic follow-ups.
NLX is the new UX... Conversations also have grammars. They have structures. They have UI elements. They're invisible.
— Aparna Chennapragada