⏱️ Estimated reading time: 3-4 minutes

I began thinking about the Hungarian Baby Name Selector app (HBNS) in 2018 to help Hungarian parents choose baby names that align with national naming laws and cultural customs, thereby easing a traditionally stressful process. By 2025, it became my testbed for AI-assisted “vibe coding” development, transforming from a prototype into a full-featured, production-ready app. Here's how it unfolded.

HBNS is a bilingual, mobile-friendly web app with:

Compliance to Hungary’s official name registration laws and customs

Real-time partner collaboration on name selection

An intuitive swiping interface categorizing names as “Yes,” “Maybe,” or “No”

Options for personalized display and filtering

Project and user account management with various authentication modes

Engaging winner selection animations and celebratory feedback

Privacy-minded analytics tracking user engagement

Clean, accessible claymorphic design for seamless device use

It focuses on making baby naming enjoyable, collaborative, and culturally sound.

<a href="https://mineknevezzelek.hu">&gt;&gt;&gt; See below for an embedded version, or find the HBNS app here. &lt;&lt;&lt;</a>

With a background in product design, I developed detailed personas, user flows, and feature prioritisation roadmaps matching Hungarian parents’ needs. AI-supported documentation (using tools like Gemini Deep Search and Claude) ensured a clear, living product blueprint pivotal for subsequent AI-assisted development.

<a href="https://www.notion.so/HBNS-Hungarian-Baby-Name-Selector-App-PRD-9f40b094ce1c4f7a89c8e9f274ae6e60">&gt;&gt;&gt; See below for an embedded version, or find the Product Requirements Document here. &lt;&lt;&lt;</a>

The AI vibe coding landscape is evolving incredibly fast. My findings represent my experiences from August 2025, specific to HBNS and my skills - the best tool today might change in weeks.

Base44&nbsp;and&nbsp;Lovable&nbsp;for their designer-friendly reputations

Replit&nbsp;for its slightly more developer-oriented setup that still aligns with my coding comfort and project complexity

These choices were deliberate, supported by professional recommendations and thorough exploration, not casual testing.

Base44:&nbsp;Enabled building around 80% of the app but AI unpredictability led to costly regressions and blocked progress.

Lovable:&nbsp;The promising initial usability suffered from the need for prompting in tiny tasks, fast credit consumption, and persistent bugs that required repeated fixes.

Replit:&nbsp;Provided strong system-level AI task management, transparent usage, and robustness, facilitating completion of a production-ready app.

Throughout, I worked thoughtfully, meticulously capturing learnings rather than rushing, ensuring a thorough evaluation.

AI agents must have broad project understanding to be truly effective.

Thoughtful control over AI helps prevent destabilization.

Transparent credit and debugging processes are essential for productivity and trust.

Human-led testing remains indispensable despite AI assistance.

Live and ready, HBNS is helping families while plans for enhancements - like real-time syncing, AI-powered recommendations, native apps, and richer analytics - promise even better experiences.

This project has been a blend of passion, learning, and sheer curiosity. I hope it inspires others to experiment with AI-assisted development and reminds us that great products come from the harmony of human insight and smart tools.

<a href="https://www.notion.so/Hungarian-Baby-Name-Selector-App-AI-Assisted-0-1-Product-Development-2782efa0eeba80aab49ad60f5f74451b">&gt;&gt;&gt; Not afraid of reading, and interested in a more detailed version of my case study? You can find it at this link. &lt;&lt;&lt;</a>

A solo passion project using AI-powered product management to build a collaborative Hungarian baby name selector app for smooth, culture-aware naming.

>>> See below for an embedded version, or find the HBNS app here. <<<

>>> See below for an embedded version, or find the Product Requirements Document here. <<<

>>> Not afraid of reading, and interested in a more detailed version of my case study? You can find it at this link. <<<

Hungarian Baby Name Selector App - AI-Assisted 0-1 Product Development

I designed and optimized OptiMonk 3.0's drag-and-drop editor and AI-powered personalization tools, streamlining user workflows for A/B testing and analytics to boost conversions for 30,000+ ecommerce brands. Collaborating with Product, Engineering, and Marketing teams, I aligned design decisions with business goals, deepening my passion for product management.

Led product-led redesign of newsfeed & onboarding for AI fintech tool, built accessible design system, implemented guest access, and crafted UX communication to drive engagement & conversions.

OKI Europe Ltd (Japanese multinational): Directed marketing and partner strategies in Hungary, aligning cross-functional teams, optimizing budgets, and driving stakeholder engagement for product success.

WPViking Ltd (Hungarian edtech startup): First hire, scaled user base to 4,000+. Shaped content, optimized user acquisition, and built WordPress sites, enhancing product adoption.

Built personal brand teaching Polish & Hungarian (A1–B2), created personalized e-learning content, led diverse groups, and honed cross-cultural communication, leadership, and user-focused design.

<b>BrightMeUp!, an AI-powered news aggregation product, was failing to 
attract and convert users due to a convoluted 30+ click first-user 
journey, unclear value proposition, and a complete lack of product-market 
fit after COVID-19 disrupted their B2B strategy.</b>

Negentropics Software, a Hungarian NLP company founded in 2015, created BrightMeUp!, an AI-based news curation app. The product used "Avatars" to deliver curated news from various sources, targeting initially financial professionals. Despite positive early feedback, the company struggled with product design, launch, and strategic direction.

The COVID-19 pandemic disrupted business relationships, causing the promising fintech project to stall. When I joined in 2021, I found a company with low UX maturity and no design system. As the first dedicated UX/UI designer, I was expected to drive strategic change.

Key challenges included resource constraints, a lack of paying users, poor user adoption, strategic missteps, operational issues, communication challenges, and the lack of a clear roadmap.

I conducted comprehensive stakeholder interviews to align on vision and metrics, followed by user research and market analysis. A product audit revealed a convoluted user journey that required over 30 clicks to deliver value, with issues including a complex landing page, desktop-only previews, and poor mobile support.

I identified three potential user groups:

We prioritized based on technical effort and business value, determining that with modest improvements, we could quickly serve broader segments (casual readers and specialists), while advanced features for professionals would require longer-term investment.

We adapted our business model to introduce three user roles:

We shifted our focus from a B2B to B2C, adopting a mobile-first approach, while maintaining our long-term goal of raising capital or monetizing the product.

I transformed the development process by implementing a structured framework inspired by the Double Diamond model, dividing work into two phases: Research/UX Audit/Roadmap Planning (1 month) and UX/UI Design/Development/Roadmap Execution (5 months).

I defined success metrics focused on converting visitors to registered users, tracking acquisition, activation, engagement, retention, and guest-to-registered conversion rates.

The improvements transformed the app into a more accessible platform with a better first-time user experience, supporting business objectives by converting more guests into registered users.

A key takeaway was the importance of advancing UX maturity. My role extended beyond design into product management, identifying high-impact opportunities and setting clear success criteria.

Looking ahead, I advocate for frequent, incremental testing as a core practice to ensure the product evolves with real user needs. This experience reinforced my ability to identify product opportunities, influence organisational processes, and bridge user experience with business outcomes.

<a href="https://www.notion.so/Reviving-a-fintech-newsfeed-app-driving-users-to-value-in-3-clicks-instead-of-30-2782efa0eeba80e583a2e9ec90ec4a53">&gt;&gt;&gt; Not afraid of reading, and interested in a more detailed version of my case study? You can find it at this link. &lt;&lt;&lt;</a>

Developed a strategic MVP roadmap for a fintech news app using PLG principles: reduced clicks to value from ~30 to 3, prioritizing immediate content access

BrightMeUp!, an AI-powered news aggregation product, was failing to 
attract and convert users due to a convoluted 30+ click first-user 
journey, unclear value proposition, and a complete lack of product-market 
fit after COVID-19 disrupted their B2B strategy.

The look and feel of the original news feed, which displays news from two selected Avatars

https://miro.com/app/live-embed/o9J_lE41Vm8=

Potential user groups and their ideal user roles according to the renewed business model

A pop-up promoting the selection of a pre-assembled Avatar package, preventing an empty newsfeed if the visitor arrives at the general newsfeed page rather than a specific Avatar page.

A pop-up promoting registration when the visitor reaches the maximum number of Avatars available for guest users

In-feed Avatar recommendations for deeper engagement

Reviving a fintech newsfeed app: driving users to value in 3 clicks instead of ~30

I am thrilled to meet Lilla! She learned the know-how of the web design industry incredibly fast. She has the perfect sense and taste for page building, selecting colors, or designing the site's structure. Her work shows that she cares about the client from her heart and does not rest until she sees that the finished piece could even be her personal website. She loves to try new things and does not shy away from big tasks. She expects to be tested. I would especially want to point out that she doesn’t shy away from asking questions, from which I also learned a lot. Long story short, if you work with Lilla, you cannot go wrong. She puts her heart and soul into the projects, and it shows in all her work. 

Bence Nagy

Lilla is an open-minded, well-prepared UX/UI designer who works in a very structured and well-organised way. She works enthusiastically and focuses 100% on her task. She always maintains a good relationship with every team member. We loved working with her.

Hungarian Baby Name Selector App - AI-Assisted 0-1 Product Development

Lilla Kisbéri-Varga