We are new, and we build exactly one thing: order replication.

QuantGrid subscribes to your broker's order-update stream and reads the same event the broker publishes: order id, status, filled quantity, product, symbol, exchange. Nothing is scraped from a rendered page, nothing is inferred from a screenshot, and nothing depends on a browser tab staying open on your laptop. If your machine goes to sleep at 10:14, replication does not.

Cost: one adapter per broker, written and maintained by hand

Zerodha, Angel One and FYERS describe the same order in three different vocabularies. Side is a string in one and an integer in another. Symbols and product codes do not match. We convert all of it into a single internal shape before anything is replicated, which is what makes a Zerodha master driving a FYERS child an ordinary case rather than a special one.

NSE:RELIANCE becomes NSE:RELIANCE-EQ · MIS becomes INTRADAY · NRML becomes MARGIN

A reconnect, a retried HTTP call and a repeated event all look identical from the outside, and any of them can place a second child order into a live market. QuantGrid runs three independent checks before an order reaches a broker: a marker key on the event, a lock held per account and symbol, and a uniqueness constraint in the database. Any one of them alone would usually be enough. Three is what makes the word usually unnecessary.

Independent layers: 3

Child quantity is the master quantity multiplied by your copy factor, floored to a whole lot, then capped by the maximum you set. When the multiplier lands between two lots we take the lower one. An exchange rejects a partial lot outright, so rounding up would either fail the order or trade size your multiplier never asked for. A slightly smaller position is the cheaper mistake, and we take the cheaper mistake every time.

Rounding: down, always

The WebSocket is the primary path because it is the fastest one. It is not the only one. If a socket drops, REST order-book polling takes over automatically and replication carries on while the connection is rebuilt, then hands back when it returns. Both paths run through the same duplicate checks, so the handover cannot double-place an order and you are not asked to do anything about it.

Fallback: REST order book, automatic, no operator action

QuantGrid runs from 8:00 AM to 5:00 PM IST on NSE and BSE trading days. Outside those hours, and at weekends and exchange holidays, the platform is taken offline for maintenance and to keep running costs down. There is nothing to replicate while the exchanges are shut, a cloud server bills by the hour whether or not anything is running on it, and not paying for those hours is part of how the price stays where it is. The decision has a consequence, and we would rather print it here than have you find it out the hard way. Place your orders while the platform is up. An order placed on the master account outside the hours above, including an after-market (AMO) order, is not replicated at all: it is already on the order book by the time the engine starts, so it is treated as history rather than as a new order, and a later fill on it does not replicate either.

Uptime: 99.9% during NSE and BSE market hours · Live status at quantgrid.in/status

Broker APIs care where a request comes from. On a shared pool your API traffic sits behind the same address as strangers, and their behaviour becomes your problem. Every QuantGrid account gets a static IP reserved to it alone. It is the headline piece of infrastructure here, and it is also the only thing that separates our two price tiers.

Shared with: nobody