A multi-broker trading platform for Indian markets
- Type
- Client project, anonymised
- Duration
- 8 months
- Team
- 2 senior and 2 junior developers
- Stack
- Python, MySQL, cloud server
- Brokers
- Zerodha, Angel One, Dhan, Jainam, MasterTrust
- Scale
- 20 accounts, the client’s 6 strategies
- Markets
- Stocks, futures, multi-leg options writing
- Outcome
- Runs on its own every session
What the client needed
The client had six trading strategies of their own and wanted every one of them to run for 20 accounts, spread across five Indian brokers: Zerodha, Angel One, Dhan, Jainam and MasterTrust. Each broker’s API works a little differently, from the daily login to the way an order is written. Doing all of that by hand, account by account, was slow and easy to get wrong.
What we built
One platform that holds the strategies and sends their orders to whichever broker each account is with. Think of a universal charger: the strategy stays the same, and only the plug changes for each broker. The client’s six strategies run on it for all 20 accounts, across stocks, futures and options.
How changes are made
- A new strategy. We add it to the platform. The broker connections stay exactly as they are.
- A new broker. We build only that broker’s connection. The strategies are not touched, so what already works is left alone.
That separation is the main design decision. It is what lets the platform grow one strategy or one broker at a time.
How we worked
Two senior developers designed the system and reviewed every important piece of work. Two junior developers built and tested parts of it under their guidance, which kept the quality high while the work kept moving. The full build took eight months.
The result
The platform runs on a cloud server and works through every session on its own, so the client does not have to start it by hand each day. New strategies can be added as the client’s needs grow.
For the broker-by-broker detail, see our Indian broker API integration page. Automating orders through an Indian broker also has to fit SEBI’s rules for retail algo trading, which our plain-English guide explains. For one system serving many accounts and brokers more generally, see multi-broker platforms and our other multi-broker platform case study, built on Interactive Brokers and NinjaTrader.
This is an anonymised account of client work. The client’s identity, the strategy logic and any trading figures have been removed, and the strategies are the client’s own. We build technology. We do not provide investment advice, we do not manage money, and we make no claim about how any strategy performs.