Website and web app development

We build systems, not just websites.

We make websites and web apps together with what runs behind them: queues, rules, integrations, payments. One of these systems takes in 20 million messages every day.

What a system we build is made of:
  1. Interfaces React · Next.js · Astro
  2. Server Node.js · NestJS · Python
  3. Data and queues PostgreSQL · Oracle · Redis 20 million messages a day
  4. Infrastructure Docker · Traefik

Measured in your browser · just now

Server response
—
First frame
—
Transferred
—

What we build

We take a project whole, from the diagram to the launch, or one part of it.

  • Web apps and platforms

    Client portals, CRMs, back offices, booking and payments, SaaS.

    Timeline: from 4 weeks

  • Backend, data, integrations

    APIs, queues, rule engines, payments, exchange with banking and accounting systems.

    Timeline: by estimate

  • Websites

    A landing page, a company site, a catalog. Fast and ready for search from day one.

    Timeline: landing page from 3 days, website from 2 weeks

  • Bots and automation

    Telegram bots, including AI-powered ones, and the routine a script can take over.

    Timeline: from 1 day

Experience

We don’t name clients. The figures and the technology are real.

Financial monitoring for a bank

The system continuously pulls customers, accounts, balances and transactions from core banking: about 20 million Kafka messages a day, which arrive in roughly 4 hours. The same business day it runs them through the rules: risk criteria and suspicious-activity indicators.

Every hit adds points to the customer’s score. Once the score is high enough, the system opens an alert, and an analyst and controllers review it. In a quarterly session about 300 rules go through 1.5 billion rows; most rules analyze transactions, which make up 95% of all records in the database.

Stack: Node.js, TypeScript, Kafka, Oracle, PL/SQL, Redis, React

  • 20M messages a day, in about 4 hours
  • ~300 rules: risk criteria and suspicious-activity indicators
  • 1.5B rows in a quarterly session
  • 95% of the records in the database are transactions
  1. Kafka
  2. Parsing
  3. Rules
  4. Score
  5. Alert
  6. Decision

Trading platform: Forex and CFD

Real-time quotes, order execution, automated trading and a back office where access rights go down to a single field.

Stack: NestJS, PostgreSQL, Redis, WebSocket, React

  1. Market data
  2. Quotes
  3. Order
  4. Execution
  5. Back office

Booking and CRM for an entertainment venue

A website in two languages, nine pricing models, card payments through two providers, a CRM for the staff, reports in PDF and XLSX.

Stack: Next.js, NestJS, PostgreSQL, RabbitMQ

  1. Website
  2. Booking
  3. Payment
  4. CRM
  5. Report

QR menu and ordering for restaurants

SaaS: a guest scans the code on the table, orders and pays without an app. Seven languages, and the kitchen sees the order at once.

Stack: Next.js, SignalR, Stripe

  1. QR code
  2. Menu
  3. Order
  4. Kitchen
  5. Payment

Telegram bots

An AI bot that analyzes video. A bot that watches for tickets to appear. Reminders and statistics for groups.

Stack: Python (aiogram), NestJS (Telegraf), Gemini

  1. Message
  2. Bot
  3. AI model
  4. Reply

Our own infrastructure

A server, Docker, Traefik, CI/CD and logs. At home, a smart home on Home Assistant.

Stack: Docker, Traefik, GitHub Actions, Home Assistant

  1. Commit
  2. Checks
  3. Image
  4. Server
  5. Logs

Process

Five steps. After each one you have a result you can check.

  1. Talk and brief We find out the task, the constraints and how you will know the project worked.
  2. Diagram and estimate We draw how the system will work and name the timeline and the cost.
  3. Development Short iterations. After each one we show a working version.
  4. Launch Domain, server, monitoring, checks of speed and search markup.
  5. Support We stay in touch: fixes, growth, keeping an eye on how it runs.

Standards and stack

What comes with every project without being asked.

  1. A speed budget

    We agree on page weight and load time and hold them with automated checks.

  2. Search markup from day one

    Metadata, structured data, a sitemap, language versions.

  3. Accessibility

    WCAG 2.2 level AA: keyboard, contrast, screen readers.

  4. Everything is yours

    We hand the code, the access and the documentation over to the client.

  5. Automation

    Checks and deploys run by themselves, not from a developer’s memory.

Stack

Interfaces
React, Next.js, Astro, TypeScript
Server
Node.js, NestJS, Python
Data and queues
PostgreSQL, Oracle, Redis, Kafka, RabbitMQ
Infrastructure
Docker, Traefik, GitHub Actions, Linux

Questions

What people ask before the first talk.

How much does a project cost?

There is no price list on the site: the price depends on the scope. After the brief we name the cost and the timeline, and if the budget is tight we suggest where to start.

How long does it take?

A bot from one day, a landing page from three days, a company website from two weeks, a web app from four weeks. We name the exact timeline after the brief.

How does payment work?

We agree on payment before we start. Large projects are split into stages: each has its own result and its own payment.

What happens after the launch?

We don’t disappear. We fix bugs, grow the product and keep an eye on how it runs, on terms we agree on.

Who owns the code?

You do. We hand over the code, the access to servers and services, and the documentation.

We have no technical specification. Is that a problem?

No. It is enough to describe the task in your own words. We draw up the system diagram and the specification together in the first step.

Tell us about the task

Leave a request and we will get back to you. A contact is enough: the rest we can discuss when we talk.

Or message us on Telegram: @exc1usive

What you need

One sentence is fine.

Telegram, phone or email.

The request comes straight to us. What happens to the data.

The form needs scripts. Without them, message us on Telegram.