Project Vectorlogo-vernice.svg

VERNICE: Neural Network for Paint Mixing

Machine learning system to predict paint formulas from specific colors in RGB format

A paint manufacturing company had 20,000 formulas in their database. Each paint uses 13 different mixtures to achieve a specific color. When a customer requests a particular RGB color, finding or creating the right formula requires time and expensive testing.

Slider Image
Slider Image
Slider Image
Slider Image

We developed a neural network-based system that predicts paint formulas starting from the desired color. The network is trained on the entire historical database and suggests the exact proportions of the 13 required mixtures.

The system is supported by an interactive dashboard that communicates with a backend based on serverless asynchronous queue management, using AWS SQS and AWS Lambda, to ensure a fluid and responsive experience.

Events on the SQS bus trigger the invocation of an asynchronous worker (AWS Lambda) that executes the artificial intelligence script and updates a database with the status and details of each executed task.

The historical database becomes predictive intelligence that accelerates formulation, reduces errors, and cuts the costs of physical mixing tests. Because even paints deserve a brain.

Machine Learning: Python + Keras for feed-forward neural network, preprocessing and data augmentation on 20,000-element database.

Backend: Node.js + TypeScript for API and worker manager, AWS SQS + AWS Lambda for queue and asynchronous worker management, DynamoDB for data storage.

Frontend: React + TypeScript for interactive web dashboard.

Monitoring: AWS CloudWatch for system usage monitoring and internal metrics, complete observability for debugging.

Card Vector 0
logo-anubi.svg

Data Warehouse: Anti-Money Laundering System

Computerized database for data storage in compliance with anti-money laundering regulations

Go to the project
Card Vector 1
HNP logo

Non così raro: Clinical Screening

Web platform for diagnostic kit logistics management in hospital clinical trials

Go to the project
Card Vector 2
logo-netech.svg

NETECH: Banking Calibration Management

UX/UI redesign of platform for anti-money laundering controls in banking environment

Go to the project
Card Vector 3
logo-wep.svg

WEP: Custom Global e-commerce

Transform a travel catalog into a machine for generating personalized experiences

Go to the project
Card Vector 4
logo-ffri.svg

FFRI: Artificial Intelligence and Fashion

AI tool for enhancing and blending historical textile patterns

Go to the project
Card Vector 5
satispay-logo

Satispay: Corporate WebApp

When one of Europe's most innovative fintechs asks you to build its new digital home

Go to the project
Card Vector 6
evenn logo

EVENN: Neighborhood Services App

Mobile marketplace for sharing neighborhood services with time-based currency system

Go to the project
Card Vector 7
autoctono

Autoctono: Wine E-commerce

The digital ecosystem that transforms the purchasing experience for the wine community.

Go to the project
Card Vector 8
logo-icp.svg

ICP - Meddle and Data Certification on Blockchain

Data validation from IoT platform directly on ICP blockchain

Go to the project
Card Vector 9
logo-chargecar.svg

CHARGECAR: Electric Charging Station Finder

Backend for mobile platform aggregating multiple electric vehicle charging providers

Go to the project
Card Vector 10
logo-hcube.svg

HCUBE - HTees Project: Sports Wearables

Mobile app for vital parameter monitoring via innovative patches applied to sportswear

Go to the project
Card Vector 11
logo-advice.svg

Agrinsight: AgriTech Platform for Remote Orchard Management

Remote monitoring and digital management for intensive orchards

Go to the project
Card Vector 12
logo-r-wtech.svg

R&W Tech: Web Dashboard for Centralized IoT Machinery Management

We gave a collective brain to 50 welding machines that weren't talking to each other

Go to the project
Card Vector 13
ima

IMA: IoT system for data collection and analysis

We transformed an intuition into a PoC that saved 90% of R&D costs.

Go to the project
Card Vector 14
Meddle Connect | Piattaforma IIoT AI-driven by Quinck

Meddle: AI & IoT at the service of production data

We leverage AI and IoT technologies to simplify and standardize the collection and interpretation of industrial data — empowering manufacturing companies with real-time production insights.

Go to the project