SEBI's August 2025 circular on algorithmic trading for retail investors came into effect in phases, with most major brokers compliant by January 2026. For the first time, retail investors can deploy automated trading strategies through broker-provided APIs without needing a Category 3 AIF or institutional connectivity. The change democratises a capability that was previously only available to proprietary trading firms and institutions.
What changed
Before August 2025: Retail investors could not use automated order-placement strategies via broker APIs. Any automated trading was technically in violation of exchange rules, even if many platforms looked the other way.
After August 2025: Brokers can register approved algorithms with exchanges. Retail clients can license or subscribe to these approved algos through the broker's platform. Each algo requires exchange approval — it cannot be a black-box system.
What retail algo trading looks like in practice
- Strategy marketplaces: Brokers like Zerodha (via Streak), Dhan, and Fyers now offer curated algo strategies — momentum, mean reversion, option selling — that retail customers can subscribe to and paper-trade or live-trade
- API access: Advanced users can build their own strategies using the broker's approved API framework, subject to risk controls (maximum order size, loss limits, etc.)
- Automated SIPs in stocks: Some platforms allow rule-based equity SIPs (buy ₹5,000 of Nifty 50 stocks on the first Monday of each month, rebalance quarterly)
The risks retail investors must understand
Backtesting overfitting: A strategy that worked over the last 5 years of data may fail going forward because it was optimised to that specific period. Always test algos on out-of-sample data.
Leverage and margin: Automated strategies in F&O can accumulate losses faster than a human can intervene. Set hard stop-losses at the strategy level, not just the trade level.
Technology risk: Server downtime, API errors, or connectivity loss can cause positions to go unhedged. Always have a manual override procedure.
Who should use algo trading
Algo trading is not appropriate for most retail investors. It requires programming knowledge, statistical literacy, risk management discipline, and continuous monitoring. For 95% of retail investors, passive index investing will outperform any self-built algo strategy after accounting for transaction costs, taxes, and time investment.