A missed birthday. A kid's schedule that shifted last minute. A group chat that never quite settles on a restaurant. This is the terrain AI mostly hasn't touched. But it’s a big part of life. So Sida Li decided to build for it.

Sida is the co-founder and CEO of Shared Context Lab, where she's building Cue, a personal AI that lives inside iMessage and group chats. Sida noticed that plenty of AI apps and services were being built for managing work. Her bet is that personal life, which is messier and less systemized, needed a solution too. 

"If it's not solving a fundamentally human need, it won't be an enduring product or solution," she says.

In this conversation, IDEO's Chief Strategy Officer Becca Carroll talks with Sida about the design decisions behind Cue, what it takes to build a company that runs on AI when things are constantly shifting, and why she thinks it’s so important to prototype your business model as well as your product. 

This is the second in a two-part series profiling founders from IDEO's Startups-in-Residence program. The first conversation is with Johannes Seemann of Sooner.

Listen on Apple, Spotify, YouTube, or wherever you get your podcasts.

Article Summary

• Why personal life is the harder, more overlooked design problem for AI

• The human-centered choices behind Cue

• How AI is changing the way she builds a business

• Treating a business model like a prototype

• Put it into practice

Why personal life is the harder, more overlooked design problem for AI

Work comes with built-in scaffolding: calendars, project trackers, inboxes. Personal life mostly doesn't. Sida's read is that AI has spent years getting good at problems that already have structure, while the long tail of everyday needs—remembering a friend's birthday, coordinating a family's schedule, planning a group dinner—is still a struggle.

The human-centered choices behind Cue

Several of Cue's most distinctive choices come down to one question: does this feel like something that’s on your side, or something that’s watching you a little too closely?

Sida built Cue inside iMessage instead of a new app, since texting already feels personal in a way opening another app doesn't. She made Cue proactive instead of waiting to be asked, on the idea that most people don't actually know everything a personal AI could do for them, though users can dial down how much Cue initiates if it starts to feel like too much. And she dropped a word most AI companies reach for by default: agent.

"We want to use language that consumers use in their lives," she says. "The jargon in AI today is confusing, and it feels cold." A conversation with IDEO's Tim Brown pushed her toward the word she uses instead: routine. "Agents have their own incentive. They have their own agenda," he told her. Cue, by contrast, is meant to act as an advocate "in your best interest," as Sida puts it.

Even the transparency built into Cue's memory, with every input being visible, editable, deletable, traces back to the same instinct: earn trust first. 

"If it's not solving a fundamentally human need, it won't be an enduring product or solution.” 

— Sida Li, Co-Founder and CEO, Shared Context Lab

How AI is changing the way she builds a business

Sida didn't know how to code when she started Shared Context Lab, but she ended up building Cue's front end herself. 

"It's kind of crazy how much has changed in the last six months," she says. The tools have moved fast enough that a founder can explore more ground solo before bringing in the team to build it, which changes what that early, scrappy phase of a company can look like.

Treating a business model like a prototype

Sida applies the same instinct to Cue's business model that she'd apply to the product itself: test it, learn from it, be willing to rebuild it. She leans on a framework from IDEO—desirability, viability, feasibility—to pressure-test ideas before committing to them. 

Not everything survives that test. Sida originally explored a version of the business built around enterprise clients. Everyone she pitched liked the idea. But she hit a real problem when she pushed further: there was no budget line inside those organizations for it. "It would be an uphill battle," she says. A consumer product turned out to be the better fit.

Pricing got the same scrutiny. Right now, it's simple by design. Sida is already innovating on several fronts at once with her product, a new category of consumer behavior, the interface, and the language she uses to describe all of it. She’s keeping pricing simple to leave space to learn how people respond to everything else first. 

Put it into practice

Don't shy away from messy problems. They're harder to systematize, which is exactly why they're overlooked and why they're often the biggest opportunity for innovation.

Use a framework to know when to pivot your business model. Sida's early business ideas got tested against desirability, viability, and feasibility. If an idea doesn't pass all three, that's a signal to go back to the drawing board, not push forward on conviction alone.

Pay attention to the words you use to describe your product. Language shapes trust before a single feature does. What you call something sets an expectation for how it's going to feel for people.

Explore more

Listen to more episodes
This is the second of two conversations profiling founders from IDEO's Startups-in-Residence program. Listen to the first, with Johannes Seemann of Sooner.

Try Cue
Join the waitlist at Shared Context Lab to try Cue for yourself.

Take a course
IDEO U's Designing a Business course — Learn how to make a business desirable for customers, financially viable for stakeholders, and feasible to build and deliver to the world.

IDEO U's AI x Design Thinking Certificate — Learn the core concepts of design thinking, and bring AI into your process to expand your perspective, generate more ideas, and unlock efficiencies.

About the speaker