Shipaton 2025 Winners
Announcing the winners of 2025 Shipaton
This year’s Shipaton brought together tens of thousands of participants from around the world, who collectively submitted 812 projects.
The results once again highlight what makes this community unique: rapid iteration, creative monetization, and a commitment to building valuable products that reach real users.
Now, as we celebrate the incredible creativity and effort that defined this year’s event, it’s time to reveal the winners of Shipaton 2025 — and for some of them, the celebration has already gone public. People have begun spotting Shipaton apps lighting up on billboards in New York:
Grand Prize: Build & Grow Award
Judging criteria
The Grand Prize recognizes the single best overall submission across all categories. Judges evaluated projects for:
- Innovation: Does the app bring a novel idea, feature, or use case to the market?
- Execution: Is the implementation polished, stable, and visually cohesive?
- Feasibility: Could the app realistically launch and scale beyond the hackathon?
- Integration: Does it make effective use of the required SDKs and APIs?
This award represents the complete package: a project that balances ambition with real-world potential.
1st Place: Payout
Payout took the top spot in Shipaton 2025 for both its technical audacity and tangible real-world impact. The app helps users discover class-action settlements they may already qualify for — transforming an obscure, paperwork-heavy process into something transparent, automated, and accessible. Millions of dollars in class-action payouts go unclaimed every year simply because consumers never hear about them, or don’t realize they’re eligible. Payout surfaces these opportunities in plain English, estimates potential compensation, and sends real-time notifications when new settlements open. In doing so, it makes the idea of ‘checking for class-action cash’ as routine as checking a bank balance — no law degree required.
What truly set Payout apart was how it was built. The entire app — design, code, and assets — was produced through AI-assisted development using Claude Code and Cursor. Not a single line of code was written by hand.
Payout’s numbers are also impressive with over 17,000 users, $30,017 in revenue, and 1,750 paying subscribers, supported by more than 500,000 impressions across social media.
#BuildInPublic Award
Judging criteria
The #BuildInPublic Award celebrates teams that shared their journey openly and used community feedback to improve their apps. Judges evaluated:
- Transparency and storytelling: How well did participants communicate their process, wins, and challenges?
- Engagement: Did public sharing generate useful discussion or inspire others?
- Learning and iteration: Did feedback from the public meaningfully improve the app’s design, features, or messaging?
1st Place: Gurwi – Learn Anything
Gurwi came from frustration with Colombia’s struggling education system and one founder’s determination to fix it. After graduating from high school in Riohacha — one of the country’s poorest regions — founder Camilo Peñalver realized how unprepared he was for adult life. Years of self-teaching led him to a pivotal insight: the real gap wasn’t motivation, but access. Education needed to be reimagined for clarity, practicality, and scale. When his video proposing an app “to learn anything” went viral with over 200,000 views, the mission became clear: build a mobile platform that teaches what traditional systems overlook.
Gurwi turns learning into an engaging, visual, and multilingual experience. Each lesson is interactive and structured around short, 10–15 minute modules that replace long videos and PDFs with a dynamic page-by-page format. Users tap through content filled with images, animations, and questions, earning points and streaks as they go. Lessons can be played aloud, paused, or adjusted for speed and are available in Spanish, English, and Portuguese. Beneath that simplicity lies an ambitious engineering effort: a custom .gurwi file format for rendering lessons, a companion content editor called Gurwi Educators, and a robust backend built with Flutter, Supabase, and PostgreSQL.
Developed by a two-person team under immense personal and financial pressure, Gurwi has already reached over 13,000 users and earned more than 1,000 app store reviews with an average 4.9 rating. Gurwi is a prime example of what you can achieve when Building in Public.
2nd Place: Echo Reminder
Echo Reminder rethinks how people capture thoughts on the go. Developer Luca wanted a faster, more natural way to set reminders — something between voice notes and task apps. Instead of tapping through menus or relying on Siri’s often clumsy transcription, Echo lets users simply speak. The app automatically parses voice input into structured reminders, making it effortless to stay organized mid-conversation, mid-commute, or mid-idea.
The app’s clean design and intuitive voice interface are powered by React Native and Expo, using Supabase for data sync and an Express backend with Whisper and GPT-5-mini for transcription and structured parsing. Despite being the team’s first iOS release, the app shipped successfully and now provides fast, accurate voice scheduling that feels native to the way people think and talk.
Beyond the technology, Echo Reminder reflects an honest builder story — one of learning in public, iterating fast, and finding a product vision through community feedback. After documenting the build daily on social media, Luca and his teammate Sydney refined the app around its core audience: busy young professionals who prefer speaking to typing.
3rd Place: Tomo Japan
Tomo Japan is both a travel companion and a personal milestone. After five years in finance, the developer taught himself iOS development with a single goal: build a Japan itinerary app. Their first attempt years ago never made it past the UI stage, but Shipaton became the moment to finish what they started. The result, Tomo Japan (‘Tomodachi’ meaning ‘friend’ in Japanese), brings that journey full circle: from learning to code to launching a polished, Swift-native app that helps others experience Japan.
Designed as a friendly, curated guide, Tomo Japan helps travelers navigate Japan’s rich but often overwhelming travel landscape. The app features handpicked destinations, detailed city guides, seasonal highlights, itineraries, and local gems that rarely appear in mainstream guidebooks.
Under the hood, Tomo Japan showcases thoughtful technical craftsmanship. Built entirely in SwiftUI, it leverages Firebase for data storage, authentication, and photos, while Supabase serves as a validation and content management layer before data is automated into Firestore. Beyond the code, the project’s success was amplified by daily #BuildInPublic updates across X and TikTok. Judges also loved the detailed and useful Medium posts the developer published as part of the project.
RevenueCat Design Award
Judging criteria
The Design Award celebrates craftsmanship and creativity. Judges looked for:
- Innovation: Fresh ideas in UX or visual systems
- Aesthetic quality: Beautiful, intuitive interfaces that spark delight
- Experience: Smooth gestures, satisfying animations, and clear feedback loops
1st Place: Dayloop: Everyday Timelapse
Dayloop turns your everyday moments into cinematic timelapse videos — no editing skills required. Whether it’s a year of selfies, a fitness transformation, pregnancy progress, or your pet growing up, the app automatically aligns portraits and compiles them into a smooth, beautiful animation. With features like Auto Face Alignment, a Ghost Photo overlay for perfect framing, and a playful timeline slider, Dayloop makes visual storytelling effortless and deeply personal.
Built in Swift and SwiftUI using Apple’s Vision Framework, Dayloop focuses on precision, privacy, and delight. Photos remain local to the device, ensuring data never leaves the user’s hands. Behind its pretty interface are countless hours spent debugging orientation bugs, managing memory, and fine-tuning image alignment to make the experience feel seamless.
Since launch, Dayloop has resonated across use cases and audiences, from fitness enthusiasts to parents to creators documenting daily life. The developer’s next focus is on growth — testing niche marketing campaigns, adding camera and export enhancements, and expanding into new languages.
2nd Place: SkillMe
SkillMe helps you learn any skill — one daily challenge at a time. From cooking to coding to handstands, the app breaks down any goal into personalized, step-by-step exercises powered by AI. Users set their goal, take a quick self-assessment, and instantly receive a tailored learning plan with daily challenges, progress tracking, and motivational reminders. Each challenge includes clear instructions, difficulty levels, and expected outcomes, making improvement feel achievable and consistent.
Built for iPhone in SwiftUI, SkillMe’s AI system dynamically generates challenges and even adapts the paywall copy to each user’s chosen goal — encouraging them with context-specific nudges like “Don’t lose your progress on becoming the next Gordon Ramsay” or “You’re one daily challenge away from turning the world upside down.”
3rd Place: PitchLab: Professional-Grade Baseball Pitch Tracking
PitchLab turns an iPhone into a professional-grade pitch tracking system — delivering analytics once limited to $20,000 radar setups. Built by a professional pitcher who spent years relying on those tools to reach the Major Leagues, PitchLab democratizes access to elite-level pitching data for athletes at every level. Using only a single camera, the app measures velocity, spin rate, spin axis, induced break, and pitch location with near-radar accuracy.
Under the hood, PitchLab blends computer vision, deep learning, and physics modeling into a streamlined, real-time system. A custom-trained neural network tracks baseballs in flight; a physics engine reconstructs 3D trajectories; and CoreML enables smooth, on-device inference at up to 60 frames per second. As a testament to its accuracy, the app is already in use by collegiate programs, training facilities, and MLB organizations.
Buzziest Launch Award
Judging criteria
This category spotlights the most creative, high-energy launches. Judges considered:
- Visibility: How effectively the team promoted their app across channels
Creativity: Use of unique stunts, videos, or storytelling that captured attention - Engagement: Measurable or anecdotal buzz from the hackathon community and beyond
1st Place: ReadHim
ReadHim was designed to make the buzziest launch of Shipaton 2025. With less than a month to ideate, build, and market their first-ever App Store release, co-founders Jay and Joseph went all-in on an idea they knew would spark conversation. The app gives women a data-driven way to decode romantic text conversations — analyzing tone, patterns, and red flags to help them understand a partner’s true intentions. By uploading screenshots, users receive a report powered by OCR and a fine-tuned GPT-OSS-120B model that detects manipulation patterns, interprets romantic intent, and even suggests what to text next.
Through viral marketing, the team grew an Instagram meme account that amassed over 5.2 million views, partnered with a TikTok influencer with over 2.3M TikTok followers, and executed a creative stunt featuring supercars and a robot dog promoting the app downtown. In just ten days, ReadHim achieved $1,100 in MRR.
2nd Place: Shutter Declutter: Photo Cleaner
Shutter Declutter turns photo cleanup into a daily ritual instead of a dreaded chore. The idea came after the developer hit the familiar ‘iPhone Storage Full’ notification, and realized that deleting thousands of photos at once was overwhelming.
Shutter Declutter makes tidying your camera roll simple and satisfying: each day, it shows you photos taken on that date in past years, and you swipe to delete the ones you don’t want — just like Tinder, but for your photos.
The app’s launch was a breakout success. Downloads surged after early grassroots marketing: social media posts, cold emails, posters around town, and even getting featured in The Verge. It reached thousands of active users and over 1,000 paying subscribers, driven largely by organic, word-of-mouth growth.
3rd Place: MemoLune
MemoLune captures the simplicity of jotting a thought as naturally as sending a message. The idea came when the developer noticed a friend using a one-person group chat just to leave memos; an act so simple, it felt like the right model for a better note-taking app. They set out to build MemoLune over 100 consecutive days, turning the process into a story itself.
The result is a lightweight, multilingual memo app that blends voice input, OCR, and chat-like design to make capturing ideas effortless. Users can speak, snap photos, or type their thoughts, with instant transcription and text extraction. Notes can be revisited via a calendar view, helping users rediscover what they were thinking on any given day.
Released live during the developer’s keynote at PyCon JP, the app quickly found its customers. Within the first week, the developer was invited to deliver four talks on the project at different events and was also scheduled to appear on a radio program to discuss the story behind the launch.
HAMM Award
Judging criteria
The Help Apps Make Money (HAMM) Award recognizes the strongest monetization strategy. Judges reviewed each project for:
- Clarity of strategy: Are revenue streams well-defined and realistic for the target market?
- Creativity: Does the app explore hybrid or unconventional monetization models?
- Financial viability: Could the approach sustain a long-term business or competitive edge?
1st Place: Vector Guard
Vector Guard turns public health data into protection, bridging the gap between research and real lives. The app transforms complex CDC surveillance data into an accessible, offline-first platform that helps users identify disease-carrying ticks and mosquitoes instantly.
The app’s core innovation is its 1:50 Justice Model, where every $2.99 premium subscription automatically funds 50 free accounts in high-risk ZIP codes — no applications, no ads, no internet required. Built in Swift, the app combines CDC open data, AI-driven image recognition, and multilingual accessibility to deliver prevention tools directly to the people who need them most. Vector Guard demonstrates how thoughtful design and fair monetization can turn public data into public good — making disease prevention not a privilege, but a human right.
2nd Place: Napkinmatic AI3D
Napkinmatic AI3D transforms casual sketches into creative magic — turning doodles on paper into digital masterpieces, interactive 3D models, and animated videos. Inspired by the universal act of sketching on napkins in cafés and classrooms, the app reimagines creativity at its most spontaneous. Users simply snap a photo of their napkin drawing, and within seconds, see it come to life through AI-powered transformations.
Built with React Native and Unity/Three.js, the app combines mobile accessibility with immersive rendering. The backend runs Python AI services for visual transformations, while RevenueCat powers a hybrid monetization system: users purchase ‘Napkin Credits’ that convert into Coins, or subscribe for monthly creative bundles.
3rd Place: Kigaru Talks
Kigaru Talks helps learners speak Japanese with confidence, not just correctness. The app was designed after years of teaching American military members stationed in Japan who could ace written tests but froze in real conversations. That insight became the foundation for Kigaru Talks: a lighthearted, judgment-free companion that helps people build fluency through real conversation.
The app uses real-time speech recognition, AI-tuned dialogue at JLPT levels, and gentle correction loops that convert mistakes into flashcards for review. Conversations adapt naturally to each learner’s ability and progress, while memory features ensure continuity across sessions. Designed for safety and personalization, Kigaru doesn’t grade or score users — it encourages practice at their own pace.
RevenueCat Peace Prize
Judging criteria
The Peace Prize recognizes technology that drives positive social impact. Judges evaluated:
- Impact: How effectively the app addresses a meaningful societal or community issue
Feasibility: Can the project realistically achieve its goals with available tech and scale? - Reach: Potential to improve lives — either for large audiences or deeply within smaller groups
1st Place: Heartbeat Hero
Heartbeat Hero turns your iPhone or iPad into a CPR coach — teaching users to perform life-saving compressions with accurate rhythm, depth, and confidence. It’s designed for everyone: learners, students, and everyday people who want to be ready when seconds matter. The app offers five training modes — Learn, Rhythm, Depth, AI Call Simulation (999/911), and a Real AED Map — all running entirely offline for reliability and privacy. Students receive full access free, while core learning tools remain open to everyone.
The project began as a WWDC Swift Student Challenge 2024 Distinguished Winner, later rebuilt from the ground up for public release. Inspired by the developer’s firefighter uncle, who saved a life using CPR, Heartbeat Hero was created to give anyone the knowledge and confidence to act. Its breakthrough innovation lies in Depth Practice — combining ARKit, IMU fusion, and adaptive filtering to measure chest compression depth with millimeter accuracy, no special hardware required. Accessibility is equally central, with voice guidance, haptic cues, torch flashes, and a Focus Mode designed to support users with ADHD, dyslexia, or reading difficulties.
2nd Place: Hearing Buddy
Hearing Buddy helps people with hearing challenges follow conversations in real time — providing accurate, private, and affordable captions powered by Apple’s newest on-device speech recognition. With accessibility at its core, it’s a simple, powerful companion that listens alongside you, transcribes speech instantly, and helps you stay connected in classrooms, workplaces, and everyday life.
The developer created Hearing Buddy for their partner, who is hard of hearing, after seeing how difficult it could be to catch every word in group conversations. Under the hood, Hearing Buddy uses the new Apple Speech framework in iOS 26 for fast, offline transcription, with CoreML-powered speaker detection to differentiate voices.
3rd Place: MoodHaven
MoodHaven helps parents track and understand their child’s moods, sleep, medication, and behavioral incidents — revealing patterns that lead to better care and clearer conversations with professionals; turning scattered notes and exhausted memories into organized insights families can rely on.
The idea began at home. The developer’s seven-year-old son has faced serious mental health challenges for years, navigating a system that often fails children who can’t yet explain what they feel. Appointments were too short. Diagnoses shifted. Medications changed faster than anyone could track. The need for clarity — for real, data-driven context — became deeply personal. Built in Swift with the help of AI tools like Cursor and Codex, MoodHaven blends thoughtful design with approachable technology to help you navigate your child’s mental health.
Best Vibes Award
Judging Criteria
This award honors creativity through the use of vibe-coding tools — AI and creative platforms that enhance development or design. Judges assessed:
- Integration and application: How well the AI tool or partner product improved workflow or user experience
- Creativity: Whether the team used the technology in unexpected, delightful, or innovative ways
- Execution: The quality and polish of the final product built with these tools
1st Place: Otter Day: Weekday Guesser
Otter Day turns an old-school mental math trick into an interactive, story-driven learning game. In just minutes, players can learn to tell the day of the week for any date in history — July 4, 1776? Thursday. Your 70th birthday? Friday.
The app was built solo, using Perplexity Pro for dialogue and visuals, KlingAI for otter animations, and ElevenLabs for voiceovers. The developer leaned on AI tools as creative collaborators; refining art, story, and UI while maintaining a handcrafted tone.
2nd Place: Dripped
Dripped is the answer to the eternal question: “What should I wear today?” Built by a self-proclaimed ‘bad dresser’, the app was inspired by frustration with daily indecision and mismatched outfits. After wasting countless mornings and relying on ChatGPT for quick color checks, the creator decided to build a personal AI stylist — one that could understand their wardrobe, suggest outfits automatically, and save time every single day.
The app is powered by SwiftUI, Metal, and a Bun backend. What truly sets Dripped apart is its unique PR-driven vibe coding workflow: open an issue GitHub tagging Claude Code, which generates a pull request implementing the feature or fix, CodeRabbit performs lint review, and last a custom GitHub Action triggers an Xcode Cloud workflow to ship a TestFlight build to test on the go.
3rd Place: MaestLog: Your Personal Symphony Journal
MaestLog is an elegant concert diary built for classical music lovers who want to preserve their listening journeys. It began with the realization that memories of live performances fade quickly. After attending countless concerts with their violinist partner, the developer found that details — dates, orchestras, pieces — were slipping away. With MaestLog, users can log their concerts, complete with dates, venues, performers, and detailed program information. Each record includes personal reflections and ratings, building a living archive of musical memories.
Starting with zero mobile development experience, the creator learned Swift, built a full iOS app, and integrated RevenueCat subscriptions — all in three weeks. Using tools like Claude Code, CodeRabbit, and Codex, every step was a collaborative learning experience between human and AI.
OneSignal Boost Award
Judging criteria
The OneSignal Boost Award honors teams that demonstrated best-in-class use of the OneSignal API. Judges assessed:
- Integration quality: Was OneSignal implemented cleanly and stably?
- Impact on experience: How did notifications improve retention, engagement, or usability?
- Creativity: Did the team employ notifications in clever or unexpected ways, such as personalized triggers or advanced segmentation?
1st Place: Cooked This: Cooking Diary and Tracker
Cooked This reimagines cooking as a measurable, motivating habit — like fitness tracking, but for your kitchen. The app helps home cooks log meals, track variety, and celebrate milestones, transforming cooking from a routine task into a rewarding practice.
Cooked This’ use of the OneSignal SDK stood out to judges: rich notifications as a behavioral design tool, encouraging users to log meals, maintain streaks, and celebrate small wins. Rather than relying on arbitrary metrics, the app rewards real progress — new dishes tried, cooking frequency maintained, or simply showing up. The result is a positive feedback loop that builds sustainable habits. The developer emphasized careful tuning of reminders and milestones to avoid fatigue, balancing encouragement with autonomy.
2nd Place: Voicetree – AI Meeting Notes for Work & Study
Voicetree was built out of a simple but universal frustration: how many times have you left a meeting, only to immediately forget what was decided? For two indie developers juggling multiple products, this daily pain point sparked the idea for a smarter, faster, and more human note-taking tool.
Voicetree is an AI-powered meeting assistant that records or uploads audio, transcribes it in real time, and transforms it into structured, editable notes. From there, users can generate summaries, blog posts, email drafts, or even chat directly with their transcripts to extract insights — delivering clarity without friction, so they can focus on conversations, not note-taking. Built in React Native and Expo with custom Swift integrations for Apple Watch, Siri intents, live activities, and widgets.
Beyond its technical polish, the team impressed judges with strong product thinking — defining a clear MVP, prioritizing user trust, and validating traction early on, with over 1,000 installs and $2,000 in revenue within weeks of launch. Judges praised its user journeys, highlighting the use of email and push notifications, experimentation with in-app messaging, and the related blog post from the team.
3rd Place: Friendy+
Friendy+ was born from a modern networking problem: we meet countless interesting people at events, but weeks later we can’t remember what they do, where we met, or why we wanted to follow up. These ‘weak ties’ are often the most valuable connections — professionally and personally — but they’re also the hardest to maintain. Friendy+ turns that challenge into an opportunity with a simple, elegant app designed to help users capture new contacts, recall details instantly, and stay in touch more intentionally.
What impressed judges most was the app’s use of OneSignal journeys and transactional messaging to provide persona-based messaging journeys that encourage follow-up, and track engagement through outcomes.
Through a six-part YouTube series, the developers shared their process, pivots, and lessons learned. Early feedback revealed that their initial audience (close friends) wasn’t the right fit, so they quickly pivoted to focus on professional acquaintances instead.
4th Place: Studient
Studient sets out to make studying smarter, faster, and a lot less painful. Built in just seven days for Shipaton, it’s an AI-powered learning companion that turns any PDF into structured study materials — complete with flashcards, quizzes, and summaries. The idea came from a simple late-night frustration: if AI can generate code and design, why not let it generate personalized study tools too?
From a technical standpoint, Studient was an ambitious experiment in vibe coding. Built with Flutter, Studient runs smoothly on both iOS and Android, while OneSignal delivers friendly study reminders. Judges praised the app’s originality, use of Journeys, and the polished user experience.
5th Place: Camp Notes
Camp Notes helps campers remember the best sites, avoid the bad ones, and share their outdoor experiences. The idea started after the developer went on a few camping trips with his wife, and realized there was no easy way to log which campsites were worth revisiting and which to skip.
With Camp Notes, users can record detailed campsite visits, including photos, notes, ratings, and attendees, while contributing to a growing, crowdsourced database of campgrounds. Built natively in SwiftUI with Firebase powering real-time sync, Camp Notes displays thoughtful design and technical craftsmanship. Judges were especially impressed with the use of omnichannel messaging for onboarding and reengagement.
Kotlin Multiplatform Reach Award
Judging criteria
This award recognizes teams that built exceptional cross-platform apps using Kotlin and Compose Multiplatform. Judges looked for:
- Technical execution: Was the app stable and feature-complete on both iOS and Android?
- Design consistency: Did the experience feel native across platforms while maintaining a unified brand identity?
- Community contribution: Did the team share their learnings, challenges, or resources publicly to support other Kotlin developers?
1st Place: Momental
Momental emerged from the creator’s search for stillness amid constant digital noise. After traveling to a monastery in Nepal to study meditation, the calm quickly vanished back home — buried under notifications, logins, and the over-engineered complexity of modern mindfulness apps. That experience sparked a clear design philosophy: meditation doesn’t need more content or community features; it needs less friction.
The result is a beautifully-minimalist meditation and focus app built on one principle: one page, one tap, nothing more. Users can meditate, fall asleep, or focus using clean timers and customizable soundscapes. Built with Kotlin Multiplatform, Momental runs natively on iOS and Android, combining AI-generated and open-source audio edited for perfect quality. There’s no login, no ads, and no onboarding.
Despite its simplicity, Momental quickly gained traction, with over 4,000 sessions recorded in its first month and an unexpected range of use cases, from teachers helping students focus to tinnitus sufferers finding relief.
2nd Place: Posturely
Posturely started with an observation: hours spent coding, working, or even solving puzzles can silently wear down your posture — and your focus, energy, and confidence with it. By combining input from your phone camera, laptop webcam, and even AirPods tilt sensors, it monitors posture in real time and gently reminds you to adjust before discomfort sets in.
The app offers real-time detection, personalized feedback, and guided exercises, all built around a friendly giraffe mascot that keeps the experience approachable rather than clinical. Users can view progress through posture streaks and reports, and even block distracting apps until exercises are completed. Built with Kotlin Multiplatform, Posturely runs across iOS, Android, and desktop with a single shared codebase. It integrates Supabase for backend data sync, and takes a privacy-first approach, keeping all posture data on the user’s device. Judges praised the developer’s focus on supporting three platforms and being very creative.
3rd Place: Steps Share
Fitness gets more fun when it’s shared — this is the fundamental idea behind Steps Share. After watching friends swap step-count screenshots in a WhatsApp group, the developer set out to build a tracker where accountability was built-in from the first tap. Steps Share turns daily movement into a friendly challenge, letting users compare progress in real time through live rings and ‘Friend Duels’.
Built with Kotlin Multiplatform, Steps Share syncs data through Firebase and connects to both Google Fit and Apple HealthKit. Currently in beta, users who connected with at least one friend increased their daily steps by 22% — proof that small doses of social motivation drive real results.
4th Place: DrawIt: A Multiplatform Draw & Guess Game
DrawIt reimagines the classic party game of sketching and guessing for the cross-platform era. Players can create private rooms, invite friends, draw clues in real time, and race to guess words while climbing the leaderboard. The experience is fast, social, and packed with small design details such crisp animations, satisfying sound effects, and a playful interface that feels at home on both mobile and desktop.
DrawIt runs seamlessly across iOS, Android, and desktop, even supporting Apple Pencil input. The team engineered a real-time backend using Firebase — leveraging Firestore, Cloud Functions, and Cloud Run — to manage game rounds, word selection, and leaderboard updates.
5th Place: ClipUGC
ClipUGC was built to solve a familiar indie-builder problem: you can ship products, but marketing — especially those high-conversion, influencer-style hook videos — feel out of reach. Agency work is pricey, and many AI tools are tuned for corporate teams with equally corporate pricing. ClipUGC pares the job down to what creators actually need.
The workflow is clear: generate or customize an AI influencer from text prompts, upload a demo video, integrate the hook, and let the app generate a polished intro video that makes people curious. UI is sprinkled with small moments of delight, like a confetti finish and hidden tap-to-unlock easter eggs. The result is a practical tool that helps indie devs, startups, and small businesses create on-brand hook videos in minutes.
Staff and Sponsors Award
Judging criteria
Reserved for RevenueCat employees and sponsor teams, this award honors the top overall internal project. Judges focused on:
- Concept originality: Does the idea stand out or solve a novel problem?
- Execution and design: Is it production-quality and enjoyable to use?
- Monetization: Does it demonstrate thoughtful or experimental revenue design?
1st Place: Crystal Abyss
Crystal Abyss reimagines the classic Columns puzzle formula through the lens of Monument Valley’s serene minimalism and Dante’s Divine Comedy’s haunting depth. Players drop stacks of shimmering jewels through the nine circles of hell, clearing matches and uncovering meditative rhythms beneath the chaos. Each level represents one of Dante’s circles — from the still calm of Limbo to the fractal frenzy of Treachery.
Built with SpriteKit and SwiftUI, Crystal Abyss merges art direction with smooth, physics-driven gameplay. Every animation, from the cascade of falling jewels to the glow of completed matches, was crafted to evoke a sense of transcendence. Behind the scenes, the project became a case study in AI-assisted development: using Claude and Codex, the developer vibe coded the entire game.
It’s anchors away for Shipaton 2025
Over the past two months, you’ve absolutely outdone yourselves. From solo builders launching their first-ever apps to seasoned developers pushing the boundaries of what’s possible with AI and monetization, the creativity and determination we’ve seen during Shipaton 2025 have been nothing short of inspiring. In just eight weeks, participants turned ideas into reality — shipping projects that were polished, original, and full of heart.
Finally, from all of Team RevenueCat: congratulations to all the winners! And thank you to everyone who contributed, mentored, or simply encouraged someone else to ship.
While it might be the end of Shipaton 2025, it’s just the beginning of your apps. We can’t wait to see where they go.
How did you find Shipaton? Did you have fun? Were there things you wish we had done differently?
We’d love to hear your thoughts. Let us know on Twitter or LinkedIn, and keep an eye out for a survey we’ll be sending to all participants to gather your feedback.
You might also like
- Blog post
New York App Week 2025: Full schedule and how to join
Discover the complete schedule for New York App Week 2025 (Oct 13–16)
- Blog post
Introducing RevenueCat MCP
Manage subscriptions through AI with RevenueCat’s new MCP server.
- Blog post
Here’s the full lineup for App Growth Annual 2025
Three keynotes, four talks, 24 workshops, competitions, prizes, and more besides. Are you signed up?