SPOT TRADING PLATFORM / EXCHANGE
PROPELOO engineers spot trading platforms — order book, matching engine, real-time market data, professional trading UI, REST/WebSocket API and the liquidity infrastructure that makes professional traders want to trade on your platform. Spot trading is the foundation product of any cryptocurrency exchange.
Professional traders choose exchanges based on order book depth (tight spreads, large available volume at each price level), API reliability (no missed WebSocket messages, low latency REST), and execution quality (limit orders that fill at the specified price, market orders with minimal slippage). A spot trading platform that performs well on these metrics attracts market makers — who make the spreads tight — which attracts more traders, which generates more volume. The virtuous cycle begins with engineering quality.
Price-time priority order book with sub-millisecond matching. All standard order types: limit, market, stop-limit, stop-market, iceberg, fill-or-kill (FOK), immediate-or-cancel (IOC), post-only.
WebSocket order book snapshots and diffs, real-time trade feed, OHLCV candles (1m, 5m, 15m, 1h, 4h, 1d), 24-hour price change, best bid/ask ticker, funding rate (if derivatives).
Desktop-optimised trading UI with depth chart, order history, trade history, order placement panel, portfolio overview and TradingView advanced charting. Keyboard shortcuts for professional traders.
Mobile-first simple trading interface — select pair, choose buy/sell, enter amount, confirm. Designed for non-professional users who want simple market access without order book complexity.
REST API for account management, order placement, balance queries and trade history. WebSocket API for real-time market data and account update streams. FIX 4.4 protocol for institutional algorithmic traders.
Admin-configured trading pairs, base/quote currency configuration, minimum order size, tick size, price filter and market-specific fee configuration.
Professional traders judge a platform in the first hour of API integration. If the WebSocket drops messages, if REST latency is inconsistent, or if the order book state ever becomes inconsistent, they leave and never return.
Professional traders build automated systems on trading APIs. A single missed WebSocket message in their order book subscription means they are trading on stale data. The WebSocket feed must guarantee delivery, reconnection must provide a full snapshot at the current state, and the REST API latency must be consistent (p99 < 50ms, not average < 50ms). Consistency and reliability matter more than peak throughput for professional use.
Axiom:
Without market makers, spreads are wide and order books are thin. With aggressive market makers (maker rebates, low latency API, co-location support), spreads tighten and depth increases, attracting more traders. The market maker programme — including negative maker fees (paying market makers to provide liquidity) — is a direct investment in the trading platform quality.
Axiom:
A professional trading interface maximises information density — full order book depth, detailed position history, keyboard shortcuts. A simple interface minimises cognitive load — hide the order book, show only current price and a simple buy/sell form. These are fundamentally different UX philosophies. Build them as separate views, not as "simple mode" vs "advanced mode" of the same interface.
Axiom:
Low fees attract high-frequency algorithmic traders who make markets. Higher fees with lower minimums attract retail users who trade less frequently. Most successful exchanges have a tiered fee structure: high-volume traders pay lower fees, creating an incentive to concentrate trading volume.
Axiom:
Impact: Maker/taker with volume tiers is the industry standard — negative maker fees (-0.01% to 0%) incentivise market making, and volume tiers reward high-frequency traders. Token-based discount adds later as a token utility mechanism.
Impact: Limit, market, stop-limit and post-only at launch — covers 90% of use cases. OCO (one-cancels-other) and iceberg orders add for professional trader segment.
Impact: VPS co-location recommendations and eventually official co-location for institutional clients. Required to attract high-frequency market makers.
Impact: TradingView Advanced Charts for professional trading interface — it is the industry standard and what professional traders expect to see. Lightweight Charts for mobile interface.
Impact: React Native mobile app for the simple trading interface. Professional trading interface web-only — professional traders use desktop anyway. Native when budget permits.
Full-featured spot exchange with professional trading UI, all order types, market maker API, REST/WebSocket API and FIX protocol.
Mobile-first simple trading app — buy/sell at market, current price display, portfolio value, price alerts and push notifications.
Brokerage model — simple buy/sell interface, no order book visible, execution at best market price, spread-based revenue model.
Spot exchange designed for specific geography — local payment methods for fiat on-ramp, local currency pairs, local language.
Exchange with integrated OTC desk for large trades — RFQ system, minimum trade size, OTC execution at negotiated price, settlement workflow.
API-only spot trading platform for algorithmic traders — REST, WebSocket, FIX protocol, co-location, low-latency order routing.
Stack: Go / Rust, Redis (order state), Kafka (event log), PostgreSQL (settlement)
Stack: WebSocket server (Node.js/Go), Redis pub/sub, TimescaleDB (OHLCV), Custom market data aggregator
Stack: React + TradingView, WebSocket client, Order placement forms, React Native (mobile)
Stack: REST API (Fastify/Go), WebSocket streams, FIX protocol (quickfix), HMAC authentication
Stack: AWS multi-AZ, Kubernetes, Cloudflare DDoS, Datadog, PagerDuty
Stack: Market maker API, Maker rebate programme, Co-location infrastructure, Liquidity analytics
Every order validated before entry into the matching engine: price within reasonable range, size within limits, account has sufficient balance, not a duplicate order ID.
API keys with fine-grained permissions (read, trade, withdraw). IP whitelist per key. HMAC signature on all requests. Key rotation without downtime.
Authenticated WebSocket connections for account data streams. Signed subscription requests. Connection rate limiting. Reconnection handling without data gaps.
Trading platforms are prime DDoS targets — competitors and manipulators both benefit from outages. Multi-layer DDoS (Cloudflare + AWS Shield), order rate limiting per API key, connection limits per IP.
Immutable event log of every order action. Real-time reconciliation between matching engine state and ledger. Any discrepancy triggers an immediate alert.
User funds in exchange wallets follow the same hot/cold architecture as described for CEX development — 80-90% cold, 10-20% hot, automated sweep, multi-sig withdrawal.
Matching engine design, market data architecture, API design, fee model.
Order book, matching algorithm, all order types, settlement. Load tested.
WebSocket server, OHLCV engine, ticker, order book streaming.
Professional web UI, TradingView integration, order placement, portfolio.
REST and WebSocket API, FIX protocol for institutional, API documentation.
Market maker API, rebate programme, co-location preparation, initial liquidity.
Security audit, load test, beta with trusted traders, market maker onboarding.
Price-time priority is the most common matching algorithm: among all buy orders at the same price, the one placed earliest gets filled first. Among all sell orders at the same price, the one placed earliest gets filled first. The best buy order (highest price) matches against the best sell order (lowest price). This is the algorithm used by virtually all professional exchanges — it rewards limit order placement timing and is well-understood by traders.
A post-only order guarantees that your order will be placed in the order book as a maker order — it will never immediately match against existing orders. If placing the order would cause an immediate match, the order is rejected instead. Post-only orders are used by market makers who want to earn maker rebates and avoid paying taker fees. They ensure the trader always provides liquidity rather than taking it.
Co-location means hosting a trading server physically within the exchange's data centre, connected directly to the matching engine over a local network connection. This reduces round-trip latency from a typical 10-50ms (internet) to sub-millisecond (local LAN). For high-frequency market makers and arbitrageurs, even 1ms of latency matters — their strategies depend on responding to market events faster than competitors. Co-location is typically offered to institutional clients at a premium.