Arihant Algo LLP

We build and run our own trading engine.

Arihant Algo LLP writes and operates a price-ladder trading system for equity, futures and commodities on Indian exchanges. It trades our own capital, and it runs on AWS in Mumbai.

A BUY ladder: entries on each price step below a fixed reference price, each closed at its own target one step higher. 1000.00 ref 998.00 996.00 994.00 bought on a step sold at its target
How a BUY ladder works. The reference price is fixed when the strategy starts. Each step down buys one tranche, and each tranche sells at its own target one step up. Illustrative prices.

The engine

One program owns every strategy, every order and every position. We wrote it to be strict about what it does not know.

Price ladders

Buys or sells in steps away from a reference price fixed at the start, and books each step at its own target.

One strategy, several brokers

A strategy can run on more than one broker account at the same time, each at its own size.

Fails safe

Missing data or an unknown state means no new orders. Every strategy has a kill switch, a daily loss limit and an exposure cap.

Checked against the broker

Every order is matched with the broker's own order book. A mismatch stops the strategy until a person resumes it.

Runs the trading day

Logs in, starts, pauses and resumes on schedule, and picks up where it left off after a restart.

Tested

2001 automated tests that run offline, and two audits of how it handles failure, in October 2026.

On AWS

Everything runs in the AWS Mumbai region, ap-south-1. This is what is live today.

Our brokers

Market dataLive prices over WebSocket
OrdersPlaced and tracked over REST

Amazon EC2, Mumbai

Trading engineEvery strategy, order and position
Trading deskWeb app for our traders, behind TLS

Stored on AWS

Amazon RDS for PostgreSQLStrategy state, every order and fill
Amazon S3 and Amazon AthenaLogs as Parquet and daily trade history, queried in SQL
AWS Secrets ManagerService secrets, read at start-up

Why Mumbai

Every order makes a round trip to the broker, so distance to the broker decides the region. A server in Europe or the US would add about 150 ms to each one.

Least privilege

The database login can read and write the app's tables and nothing else. The server's role can only write its own logs to S3. Broker logins are stored encrypted.

About

Arihant Algo LLP is a limited liability partnership in India. We trade our own capital with software we write ourselves.

The engine is for our own use. We do not sell it, and we do not offer investment advice or trading services to anyone.

Designated partners
Sanjay Jain, Pooja Jain
Incorporated
23 December 2016
Based in
Delhi, India
LLPIN
AAI-0957

Contact

Write to us at sanjain40@gmail.com.

Registered office
House No. 20, Parking Floor, PK1, Shalimar Bagh, Mothers Pride School, Delhi 110088, India