Connecting code to the market

Automation is where a strategy meets reality: your code has to talk to a broker's servers reliably, all day, without a human watching. These pages explain the plumbing — REST and WebSocket APIs, authentication, rate limits, error handling, retries and scheduling — using broker-neutral concepts so you can build automation that is robust rather than merely working-on-a-good-day.

APIs & Automation: Trading automation connects your program to a broker through an API. REST APIs handle request/response actions (place order, fetch positions); WebSockets stream real-time data and updates. Robust automation requires secure authentication (API keys/tokens), respecting rate limits, handling API errors gracefully with sensible retry logic and idempotency, and reliable scheduling — so the system keeps running correctly even when the network, the broker or the data misbehaves.

Broker APIs

Connectivity

A broker API is a programmatic interface a broker exposes so that software — rather than a human clicking a screen — can fetch market data, read acco…

→

REST APIs

Request/response

A REST API is a request/response interface over HTTP in which each call asks a named endpoint to read or change a resource and returns exactly one st…

→

WebSockets

Streaming

A WebSocket is a persistent, bidirectional connection over which a broker pushes real-time data — market ticks and order-status updates — to your pro…

→

Authentication

Security

Authentication is how a broker API verifies that a request genuinely comes from you and is authorised to act on your account, typically through API k…

→

Rate Limits

Throttling

A rate limit is a broker-imposed cap on how many API requests you may send within a time window, enforced to protect their infrastructure, and exceed…

→

API Errors

Reliability

An API error is any response indicating a request did not complete as intended, and handling them means classifying each as transient or permanent, r…

→

Retry Logic

Reliability

Retry logic is the disciplined re-attempting of failed requests using exponential backoff with jitter and a retry cap, combined with idempotency keys…

→

Event Streaming

Streaming

Event streaming is the continuous consumption of a sequence of events — ticks, fills, signals — where the consumer must handle ordering, at-least-onc…

→

Scheduling

Orchestration

Scheduling is the automated triggering of trading tasks at precise times using cron or a scheduler, made market-aware through trading-hours and holid…

→

Automation Workflows

Orchestration

An automation workflow is the orchestrated end-to-end daily cycle of an automated trading system — pre-market preparation, the live trading loop, pos…

→

Frequently asked questions

What is a broker API?

A broker API (Application Programming Interface) is the set of endpoints a broker exposes so that programs — not just humans clicking a screen — can fetch market data, check positions and funds, and place, modify or cancel orders. In India, brokers such as Zerodha (Kite Connect), Angel One and others offer APIs; you must use exchange- and broker-approved access for automated trading.

What is the difference between REST and WebSocket APIs?

A REST API works on a request-response model: your code asks for something (place an order, get holdings) and gets one reply. A WebSocket keeps a persistent connection open and streams data continuously — live ticks, order updates — pushing them to you as they happen. Trading systems typically use REST for actions and WebSockets for real-time data.

How do I handle API errors in a trading system?

Expect them. Distinguish transient errors (timeouts, rate limits, 5xx) from permanent ones (bad request, insufficient funds), retry transient failures with exponential backoff and a cap, make order placement idempotent so a retry cannot double-send, and escalate to alerts or a kill switch when errors persist. Never assume an API call succeeded without confirming the resulting state.