Tempo - Cohort 17

Tempo Labs is a company that helps startup founders build their products using AI. They provide a dashboard where founders can create designs, preview code, and request changes easily. It’s like a smart workspace that makes product development faster and more efficient with AI-powered tools.

We built an AI Copilot that helps startup founders go from idea to execution — fast. By turning messy thoughts into structured PRDs and dev-ready tasks, we tackled the “blank page” problem head-on. Through deep user research, rapid prototyping, and tight design-engineering collaboration, we created a tool that gives founders clarity, confidence, and momentum

We designed & shipped AI Agents 0-1!

We worked closely with the YC-founders to bring this to life

Created a super intuitive design specs for handoff

Why we're excited

Working on Tempo Labs was one of those projects that really reminded us why we love design. The challenge was clear — founders often get stuck at the very start, staring at a blank screen, unsure how to go from an idea to something real. We set out to change thatOur goal was to create an AI Copilot that helps founders build faster by turning raw ideas into structured PRDs and breaking them down into tickets ready for engineering. It felt exciting to shape a tool that not only saves time but actually boosts confidence and momentum for early-stage teams. What made it even cooler? We launched the PRD generation feature with a clean handoff to devs, and the product hit #3 on Product Hunt. Being part of that end-to-end journey — from big vision to execution — was super fulfilling.

The how

01
02
03
THE BEGINNING | Our first 3 weeks
In the first few weeks of the project at Tempo Labs, our focus was on laying a solid foundation through user research and smart scoping. We kicked things off by defining clear UXR goals — we wanted to understand how early-stage founders currently build MVPs, where they get stuck, and how comfortable they were with AI-generated outputs in product workflows. we interviewed startup founders, PMs, and early-stage teams to get a sense of their mental models. A big insight was that most struggled with translating ideas into structured requirements — PRDs were often skipped or hastily done, and there was a ton of context lost between ideation and execution. That insight shaped everything we built. With the research in hand, we mapped out the core user journey and scoped down our MVP to focus on the highest-impact flow: turning a raw idea into a well-structured PRD, with the ability to generate dev-ready tasks. This allowed us to stay lean but laser-focused, and gave us a clear design and build path for the next sprint.
ideate, constrain, & test
Weeks 4 to 7 were all about rapid ideation, testing, and iteration. I started by prototyping a few core flows — how founders input ideas, how the AI generates PRDs, and where users might want control. I ran usability tests with early-stage founders and PMs, gathering quick feedback on what felt helpful vs. overwhelming. One big insight: users wanted more transparency and control over what the AI was doing, not just a black-box output. Based on that, I refined the flow to include more guided steps and editable outputs. I worked closely with engineering to align on feasibility, adjusting designs in real time. This tight loop helped us stay fast, focused, and user-first.
innovate, spec, & ship
In the final stretch, my focus shifted to polishing the experience, spec’ing out final components, and pushing toward launch. With the core flow validated, I explored ways to elevate the product — like smart defaults, inline editing, and contextual AI prompts to guide users through ambiguity. I documented detailed specs for each interaction, edge case, and state, ensuring engineering had everything they needed to build with confidence. I also partnered with them daily to QA and fine-tune the handoff — catching UX gaps early and making real-time tweaks as we went. We shipped the PRD generation feature smoothly, and it was amazing to see it resonate — founders felt empowered, and the product hit #3 on Product Hunt. It was the kind of build where design, engineering, and user needs were all in sync — and that made it special.

The what - impact & shipped

Stay organized and manage all your PRDs in one place
What is the future of AI native interface and how do we bridge the gap?

• Designing workspaces that morph based on context vs static documents

• Canvas adapts in real-time to your workflow - not just another doc editor

• Content self-organizes based on relationships vs manual folder structures

• Information flows between blocks naturally, like liquid vs rigid containers

• Elements reshape and connect based on AI understanding of your work

• Built for flow state - content appears where needed, then vanishes
Create PRDs your way—AI-powered or manual, users are in control.
Founders can skip the sprint board and start development instantly.
Designed a comprehensive sprint board to Plan, track, and let AI tackle your sprint tasks

Some of our team memories

Our AI Copilot gained significant traction, ranking #3 on Product Hunt!
In the news everywhere as one of the most anticipated YC backed AI products.
Featured on Forbes 30U30

Apply for the next cohort

apply
Applications open april 25th