The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
What if the biggest risk in a visa application isn't a missing document but a contradiction between documents that nobody noticed?
That was the problem I needed to solve for VisaGuide, a platform working with a visa agency based in Hungary.
The business wanted applicants to upload their visa documentation, identify potential inconsistencies, understand what needed attention, and receive an Embassy Readiness Report before submission.
Rather than expecting one AI model to handle everything, I broke the problem down and assigned each tool a specific responsibility.
Here was my approach 👇
1. Azure Document Intelligence —> The Extractor 📄
First, I needed to turn uploaded documents into structured data.
The OCR and document extraction layer processes the documents and extracts relevant fields that can be compared across different sources.
2. ElevenLabs —> The Interview Simulator 🎙️
Documents alone don't always provide the full picture.
I included ElevenLabs for a virtual interview simulation, creating another way to collect information from applicants and complement the data extracted from their documents with option to switch to a manual progressive form.
3. Gemini —> The Auditor 🧠
This is where the different pieces come together.
Gemini receives the extracted document data alongside the information collected through manual forms or the interview workflow. The goal is to analyze the combined information, flag potential discrepancies, and produce actionable recommendations.
The output is an Embassy Readiness Report containing:
A readiness score.
Potential inconsistencies and issues requiring review.
Recommendations on what needs to be corrected, clarified, or aligned.
The architecture
User-uploaded documents → Azure Document Intelligence → Structured document data
Applicant information → Manual forms / ElevenLabs interview → Collected data
Structured document data + collected data → Gemini → Readiness score + issues + recommendations
What interests me most about this solution isn't simply the use of AI. It's the orchestration: combining document intelligence, conversational data collection, and AI-powered analysis into one business workflow.
Because building useful AI products isn't just about choosing the most powerful model.
It's about knowing which problem each tool should solve, what data should flow between them, and how to turn the output into something people can act on.
The objective is to help applicants and visa professionals catch avoidable documentation inconsistencies earlier not to guarantee visa approval or replace professional judgment.
I'm curious: if you were architecting this system, what would you add to make the analysis more reliable rule-based validation, confidence scoring, human review, or all three?
#BackendEngineering #AIEngineering #SystemDesign #DocumentIntelligence #SaaS
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
MOTIF — From Words to Pixels to Code
What if you could see your dream website before writing a single line of code, then turn that vision into something real?
That's why I built Motif, an AI-powered visual design and website creation studio, built on Base44.
We turn words into pixels. Then we turn those pixels into real HTML and CSS.
Most AI website builders take a prompt and immediately generate a website. You get what the AI decides, and often spend ages prompting, regenerating and fighting to get something that resembles what you originally imagined.
I wanted to change that.
Motif puts visual creativity before code.
You start by describing your idea in plain English. Motif explores different visual directions, generating concepts that let you actually see what's possible before committing to a design.
Once you find something you love, Motif transforms that approved visual into a real, responsive website built with HTML and CSS. Not a flattened image pretending to be a website, but actual editable elements.
And that's where Motif becomes more than an AI generator.
Inside the visual studio, you can select elements, change typography, colours, spacing, imagery and layouts, preview desktop and mobile versions, and refine the design without starting over.
One of my favourite features is Shuffle, which can dramatically reimagine a design's composition and styling, allowing you to explore new creative possibilities without writing another lengthy prompt.
I also wanted people to understand what Motif could do before signing up or waiting for AI generation. So I built a product-focused homepage featuring real application footage, an explainer video and editable sample projects visitors can explore.
Why I built it
I've spent years working across software development, 3D design and interactive experiences, and I've always been fascinated by the gap between having a creative vision and actually bringing it to life.
AI has made building software more accessible, but great design still requires exploration, experimentation and creative control.
I believe AI should expand your imagination, not replace your creativity.
Motif is my attempt to bring those worlds together: the freedom of visual design, the intelligence of AI and the flexibility of real web development.
It's designed for entrepreneurs, developers, designers and anyone who's ever had a great idea but struggled to make it look the way they imagined.
Built with Base44, Motif demonstrates how far a creative application can go beyond traditional no-code tools.
Motif isn't about generating another template. It's about making something unmistakably yours.
Imagine it. Shape it. Make it real.
#Base44Challenge
Great idea, great output Mellis - goated
AI-powered recruitment is powerful. But can recruiters truly trust it?
Meet HireWisey, an intelligent recruitment platform combining automated CV screening, AI interviews, and candidate matching.
Our biggest UX challenge? Building trust in AI.
We focused on making every AI-driven decision transparent, understandable, and easy to review, giving recruiters the confidence to rely on AI while staying in control.
Because great AI experiences aren’t just about automation. They’re about trust.
What do you think of the design?
Yage Collections, a collections platform for small business 💳
This one started with a simple question: how do small businesses actually get paid?
After 30 interviews with owners and finance teams, the answer surprised me. Payments weren’t the problem; everything around them was. People were reconciling bank accounts by hand every week, their customers didn’t trust unbranded payment links, and juggling currencies was eating hours. So instead of building another payment tool, I designed a hub around the moment money comes in: one dashboard for every account, every branded payment link and every currency, a QR link a customer can pay in four steps, and nothing more than two clicks deep.
Every flow was designed twice, giving 160 screens across desktop and mobile. I then took it all the way into code and built the front end myself, which taught me more about my own design decisions than any handoff ever has.
Full case study on my Profile
Let’s talk.
The research behind this is impressive! Bringing payment links, reconciliation, and multiple currencies into one simple platform solves real everyday challenges for small businesses.