Heywa: AI Search Shouldn't Be a "Text Wall"—This London Team is Redefining Interaction with Visual Stories

2026-03-05 | ProductHunt | Official Site | Company

30-Second Quick Judgment

What this app does: You ask a question, and instead of a pile of text, it generates a "visual story" that you can tap, swipe, and explore through branches. Simply put, it turns ChatGPT's wall of text into an Instagram Stories-like experience.

Is it worth watching?: Yes. Not necessarily because of how perfect it is right now, but because it addresses a real pain point—AI capabilities are already powerful, but the presentation is stuck in 2010. Led by former Google/Trainline executives and backed by a $5M seed round led by Cherry Ventures, this isn't just some random side project.

Three Questions: Is This for Me?

Is it relevant to me?

Target User: Everyday people who use their phones to search for things, make comparisons, or find inspiration. Not developers, not enterprise users—just you and me.

Am I the target?: If you find yourself in any of these scenarios, you are:

• Searching for "what to eat tonight" and don't want to read a 2,000-word article.
• Comparing two travel destinations and are tired of opening a dozen tabs.
• Wanting to understand a new concept (like "what is intermittent fasting") without ChatGPT writing a thesis for you.

When would I use it?:

• Cooking inspiration → Swipe through recipe cards; it's more intuitive than a text list.
• Travel planning → Compare destinations side-by-side with images on every card.
• Learning new things → Dive deeper layer by layer without having to organize the info yourself.
• Shopping decisions → Visualize product comparisons.

Is it useful to me?

ROI Judgment: Free + Zero learning curve + Solves a real pain point = Worth 5 minutes to download and try. The worst-case scenario is just deleting an app.

Is it fun to use?

Delight Points:

• Sense of Speed: Results appear within 2 seconds of asking, much better than waiting for ChatGPT.
• Visual Impact: Every card features images or even short videos. Searching for "best coffee shops in London" shows a rotating image gallery.
• Sense of Exploration: Tap a card to go deeper; it’s as addictive as scrolling through TikTok.

The "Wow" Moment:

"Choosing paint colors? Browse palettes visually. Comparing trip destinations? See options side by side. No endless scrolling. No walls of text. Just answers you can actually explore." — @anujcodes_21 (Tweet with 59 likes)

Real User Feedback:

Positive: "Someone had to change the current standard. Most humans have visual minds, the current interfaces of the big providers seem backwards thinking!" — ProductHunt Community User

Anticipation: "ChatGPT gives you walls of text. You scroll. You skim. You open five more tabs. Heywa launches today with a better way." — @anujcodes_21

For Independent Developers

Tech Stack

• Frontend: Native iOS app (requires iOS 26.0+), also supports macOS (M1+) and Vision Pro.
• Backend: Proprietary Generative UX (GenUX) engine, built from first principles. This isn't just an API wrapper; it's a custom-built interface generation system.
• AI/Models: Specific models not disclosed. However, co-founder Martin comes from Google AI Search, and the team has experience with large-scale AI systems. Content pulls real voices from communities like Reddit and TikTok.
• Infrastructure: Headquartered in London, 12-person team.

Core Functional Implementation

Heywa's core isn't "better AI answers," but "better AI presentation." It created a middleware (GenUX Layer) that sits above the "AI model + data" and below the "specific application."

When you ask a question, this layer:

1. Identifies the question type (Travel? Health? Philosophy?).
2. Selects the most appropriate visual template (Comparison cards? Image gallery? Hierarchical exploration?).
3. Generates a structured, interactive story in real-time.
4. Dynamically adjusts subsequent content based on your taps.

Key technical challenge: How to make a system generate high-quality visual stories for "almost any question." The team's approach starts with a constrained canvas—the "story format"—and expands from there.

Open Source Status

• Is it open source?: No. There are no Heywa Labs-related repositories on GitHub.
• Similar open-source projects: Vercel's json-render does Generative UI; the philosophy is similar, but the implementation direction differs.
• Difficulty to build yourself: High. The core difficulty isn't the AI model call, but "how to map any question to the right visual template." Expect it to take 3-4 people and 6+ months to build an MVP.

Business Model

• Monetization: Not yet announced. Currently completely free and in the user acquisition phase.
• Possible directions: Similar to peers, likely a Freemium model (basic free + paid premium features) or B2B licensing of the GenUX engine.
• User base: Not disclosed, but already available on the App Store in the US, UK, and Canada.

Giant Risk

Medium-high. Google has the capability (and motivation) to add similar visual interactions to search results. In fact, Google AI Overview is already moving in this direction. But Heywa's bet is: giants can't execute this "from-scratch mobile experience" well because they are held hostage by existing search paradigms. This logic is similar to how Arc Browser challenged Chrome—it might not win, but it has a chance to build a moat in niche scenarios.

For Product Managers

Pain Point Analysis

• What problem does it solve?: The presentation of AI search results is terrible. ChatGPT writes you a thesis, Google gives you a list of links, but you actually just want to "quickly look, compare, and pick one."
• How painful is it?: Medium-high frequency; a "nice-to-have" that feels like a "must-have." It’s not "die without it," but it's the kind of experience you can't go back from. As a PH user put it: "Most humans have visual minds."

User Persona

• Primary User: 25-40 year old mobile users, accustomed to the visual interactions of TikTok/Instagram, and dislike reading long-form text.
• Secondary User: People who need to make quick comparative decisions (choosing a restaurant, a travel destination, or a plan).
• Use Cases: Cooking inspiration, travel planning, fitness knowledge, DIY tutorials, personal growth.

Feature Breakdown

Competitor Differentiation

Key Takeaways

1. "Generative UX" Concept: Don't just generate content; generate the interface itself. This mindset can be applied to any AI product.
2. Constrained Canvas Design: Starting from a limited canvas like the "story format" reduces the complexity of a "universal interface."
3. Mobile-First: Don't build for Web first and then adapt; design starting from finger-based interactions.

For Tech Bloggers

Founder Story

• Milena Nikolic: Former CTO of Trainline, managed a tech team of 400+. Spent over 10 years at Google, eventually becoming Engineering Director for the Google Play Developer Ecosystem. A prominent female tech founder.
• Martin Juan Sturla: Former Senior Engineering Lead at Google, worked on Google AI Search and the Play Integrity API.
• Why build this?: In Milena's words—"We spend hours every day in beautiful apps, but the moment we need to learn or make a decision, we're back to blue links and walls of text." This disconnect was her starting point.

Story Angle: A CTO who managed 400 people leaves a stable big-tech job to lead 12 people in challenging Google Search's interaction model. Plus, her co-founder is literally from Google AI Search—they don't lack search knowledge; they just think search needs to change.

Controversy / Discussion Angles

• Angle 1: Is AI UX being ignored? ChatGPT has been hot for two years, but the interface is still just a chatbox. Models are getting stronger, but UX has barely changed. Heywa bets that "the next battleground is UX, not models."
• Angle 2: Visual vs. Text—which is the future of AI search? Perplexity bets on text citations, Heywa bets on visual stories, and Arc Search bets on web summaries. Which of the three paths is right?
• Angle 3: Can a small team challenge a giant's interaction paradigm? 12 people vs. Google's search team of thousands. This isn't a tech war; it's an experience war.

Hype Data

• PH Ranking: 183 votes (launch day).
• Media Coverage: Exclusive coverage by Sifted, plus TechFundingNews, StartupMag, and other European tech media.
• Twitter: Promotional tweet by @anujcodes_21 received 59 likes, 29 retweets, and 6,400 views.
• Funding News: The $5M seed round is a topic in itself.

Content Suggestions

• Suggested Angle: "ChatGPT's interface hasn't changed in two years—this company wants to fix that."
• Trend Jacking: Tie it into the AI search wars (Google AI Overview vs. Perplexity vs. ChatGPT Search) and introduce the narrative of a "fourth possibility."

For Early Adopters

Pricing Analysis

Note: The product is in an early growth stage; free access is likely temporary. Use it while you can.

Getting Started Guide

• Setup Time: 2 minutes.
• Learning Curve: Extremely low—if you can use TikTok, you can use Heywa.
• Steps:
  1. Search "heywa" on the App Store to download (Currently iOS only, available in US/UK/Canada).
  2. Enter any question (e.g., "What should I eat tonight?").
  3. Swipe through cards and tap on what interests you to explore further.
  4. Each tap generates new branching content.

Pitfalls & Complaints

1. High System Requirements: Requires iOS 26.0+; won't run on older phones.
2. Regional Restrictions: Currently only live in the US, UK, and Canada.
3. Very Early Stage: The team says they are "early and very open to feedback," meaning bugs and imperfections are expected.
4. No Reddit Discussion: Almost no community discussion yet; if you hit a problem, you'll have to give feedback directly to the team.

Security and Privacy

• Data Storage: Not explicitly stated, but as an AI search tool, it will collect query data.
• Privacy Policy: Check the privacy labels on the App Store page.
• Security Audit: Not disclosed.

Alternatives

For Investors

Market Analysis

• Generative AI Market: $22.1B (2025) → $324.7B (2033), CAGR 40.8% (Grand View Research).
• UI/UX Market: $1.98B (2025) → $11.66B (2031), CAGR 32% (Mordor Intelligence).
• UX Services Market: $8.8B (2026) → $77.18B (2034), CAGR 31.2% (Fortune Business Insights).
• Key Prediction: By 2030, 90% of user interfaces will use AI to customize experiences.
• Heywa Positioning: At the intersection of Generative AI and UX/UI—an overlap of two high-growth markets.

Competitive Landscape

Timing Analysis

• Why now?:
  1. AI model capabilities are strong enough; the bottleneck has shifted from "can it answer" to "how to present the answer."
  2. 2026 is considered the inaugural year of Generative UI (GenUI); static interfaces are becoming obsolete.
  3. Mobile AI experiences significantly lag behind desktop, leaving a clear experience gap.
• Tech Maturity: Multimodal AI, real-time generation, and visual content processing are all ready.
• Market Readiness: Users have been educated by TikTok and Instagram Stories on "swipe-based content consumption"; the mental model is ready.

Team Background

• CEO/Founder: Milena Nikolic — Former CTO of Trainline (400+ person team), 10+ years at Google (Engineering Director level).
• CTO/Co-founder: Martin Juan Sturla — Former Senior Engineering Lead at Google (AI Search, Play Integrity API).
• Core Team: 12 people from Google, Apple, Snapchat, and Trainline.
• Characteristics: Female founder, diverse team culture, balanced engineering and design talent.
• Track Record: Core team members have built consumer platforms serving billions of users.

Funding Status

• Raised: $5 million (Seed round, February 2026).
• Lead Investor: Cherry Ventures (Top-tier European early-stage VC).
• Follow-on Investors: Openseed, Pareto, Plug & Play, Ventures Together, and angel investors.
• Valuation: Not disclosed (listed on PitchBook).
• Use of Funds: Product R&D, team expansion (Engineering + Design + Growth), and internationalization.

Conclusion

Heywa isn't building a "better AI answer"; it's building a "better AI experience." This direction is correct, but whether they can succeed depends on their ability to build a sufficient experience moat before the giants react.

Resource Links

2026-03-06 | Trend-Tracker v7.3