In production

nTerminal

A trading terminal built for the Indian market.

nTerminal: a strategy built from connected nodes

nTerminal: a chart with trade markers, a trades table and the N-ARCH chat

nTerminal: an option chain with a strategy panel and historical paper replay

Describe a strategy

The problem

Retail traders in India lack a fast, purpose-built terminal tailored to local market structure.

What we built

A native desktop trading terminal built around Indian market data and workflows. Describe a strategy or build it as a flow, backtest it trade by trade, then test it on an option chain with paper replay, all backed by a data platform we run ourselves.

What it does

Strategies built by chat or as a flow
Backtests with every trade shown
Option chain with paper replay
Charts, risk and portfolio

Built for

Professional traders and portfolio managers working in Indian markets.

How we built it

A native desktop client, built without Electron, backed by an authenticated server platform for market data, history and scheduled ingestion.

01What made it hard
- Live market data for many traders without duplicating broker connections or data ingestion
- Orders and broker logins that must stay on the trader's own machine, never on the server
- Data ingestion that picks up where it left off after any interruption
02How it's built
A native desktop client (Python with PySide6 and Qt Quick, no Electron) talks to a FastAPI server over REST and WebSocket. API workers are kept separate from the single broker sessions and scheduled jobs, so adding workers never duplicates provider connections or ingestion. Data lives in PostgreSQL with TimescaleDB, with Redis alongside, and an operator console runs the service.
03How we delivered it
The server owns authentication, market-data relays, historical queries and scheduled ingestion. Its broker accounts are data-only: order execution and personal trading records stay in the desktop client. The platform is containerised with Docker Compose, with health checks, backups and continuous integration.
04What shipped
- Live, with access on request
- A shared charting engine used by the desktop client
- An operator console for service status, ingestion and maintenance

What it runs on

Python
PySide6
Qt Quick
QML
FastAPI
PostgreSQL
TimescaleDB
Redis
WebSockets
Docker

Want something built like this?

The team that built nTerminal builds for clients too. Tell us what you have in mind.

Book a call Write to us

How we built it

A native desktop client, built without Electron, backed by an authenticated server platform for market data, history and scheduled ingestion.

01What made it hard

Live market data for many traders without duplicating broker connections or data ingestion
Orders and broker logins that must stay on the trader's own machine, never on the server
Data ingestion that picks up where it left off after any interruption

02How it's built

A native desktop client (Python with PySide6 and Qt Quick, no Electron) talks to a FastAPI server over REST and WebSocket. API workers are kept separate from the single broker sessions and scheduled jobs, so adding workers never duplicates provider connections or ingestion. Data lives in PostgreSQL with TimescaleDB, with Redis alongside, and an operator console runs the service.

03How we delivered it

The server owns authentication, market-data relays, historical queries and scheduled ingestion. Its broker accounts are data-only: order execution and personal trading records stay in the desktop client. The platform is containerised with Docker Compose, with health checks, backups and continuous integration.

04What shipped

Live, with access on request
A shared charting engine used by the desktop client
An operator console for service status, ingestion and maintenance