Skip to content
ServicesPortfolioAboutContact Let’s talk

Projects in production

Web apps, backends and automations: what we solved and the results we measured.

Featured, Web app

−60%

missed appointments

Reservio

Online booking platform for beauty centers and medical practices: multi-staff calendar, prepayments, automatic email and SMS reminders, admin panel with booking analytics.

ToolsNext.js, Node.js, PostgreSQL, Stripe, Vercel

Challenges and results

Challenges

  • Optimistic locking on simultaneous slot bookings
  • Partial payments and refunds with Stripe
  • Multi-channel reminders with retries on failure

Results

  • Missed appointments down 60%
  • 4 healthcare and beauty businesses in production
Web app

0

profiling cookies or tracking tools

Aedia

Website for Aedia, a Milan-based short-term rental management company: it presents the services for owners and the two fee models, the homes under management, a sample monthly statement and the form to request a free valuation.

ToolsHTML, CSS, JavaScript, Formspree, Netlify

Visit aedia.it (opens in a new tab)
Challenges and results

Challenges

  • Explaining two fee models clearly: commission or fixed rent
  • Showing transparency with a sample monthly statement
  • No profiling cookies and no traffic analytics

Results

  • Live at aedia.it
  • Valuation requests through a form, with no server to maintain
Backend and APIs

12 min

to reconcile, down from 4 hours

DataBridge

Sync middleware between a legacy AS400 system and a modern CRM. It transforms, validates and reconciles more than 50,000 records a day with no data loss.

ToolsPython, FastAPI, PostgreSQL, Redis, Docker

Challenges and results

Challenges

  • Parsing an undocumented AS400 format
  • Partial transactions with automatic rollback
  • Zero downtime during the live migration

Results

  • 50,000+ records a day, 0 errors in 6 months
  • Reconciliation from 4 hours to 12 minutes
Automation

8 hrs

saved every week

InvoiceBot

Automated handling of incoming invoices for an accounting firm: reading emails, extracting data from PDFs, filling the spreadsheet and sending a weekly summary on Telegram.

ToolsPython, Gmail API, Google Sheets, Telegram API

Challenges and results

Challenges

  • OCR on very different PDF layouts
  • Automatic classification by category
  • Handling duplicates and already processed invoices

Results

  • 8 hours a week saved by the staff
  • 200+ invoices a month with no manual work

Have a project in mind?

Tell us about it in a few lines: we reply within 24 hours.

Let’s talk