Trading bots: guides to how they work, how they are built and what they cost
Source-based guides to what a production trading bot consists of, whether bots work, which rules people automate, what the frameworks do, what the work costs and how to choose who builds it.
Short answer: A trading bot is software that applies written trading rules through an exchange API without manual clicks: it reads market data, applies the rules, checks risk limits and sends orders. A well-built bot executes rules reliably; it does not make a strategy profitable, and automated trading can lose money.
These guides follow the order a project runs in: the parts of a production bot, what regulators and research say about whether bots work, the rule types people automate, AI claims, the frameworks people start from, then the budget and how to check a developer before you hire one.
None of them recommends a strategy, a platform or a vendor. Examples use hypothetical numbers, and the rules that apply depend on your country and your exchange's terms.
Fincode sells no ready-made bot or bot subscription: it builds custom software from rules you have written and does not supply strategies. If you are comparing automated trading software, the crypto trading bot price guide explains how subscriptions, credit plans and self-hosted licenses are priced.
- Start here
- Do trading bots work, and are they legal?
- Rules people automate
- AI and trading bots
- Open-source frameworks and libraries
- Ready-made bots and platforms
- Cost and choosing a developer
- When a bot misbehaves
- At a glance
- Calculators
Start here: how a trading bot is built
The parts of a production bot, how a project moves from written rules to a small live start, and where bots usually fail.
- Trading bot development: architecture, process, pitfallsUpdated November 5, 2026 · 14 min readTrading bot development explained: the parts of a production bot, build vs buy, the spec-to-live process, common failure modes and what to ask a developer.
- Trading bot architecture: events, state and crash recoveryNovember 6, 2026 · 13 min readTrading bot architecture below the component list: event log, reconcile on start, backpressure, a separate risk process and a crash test you can run.
- Trading system development: write rules a coder can buildNovember 6, 2026 · 13 min readTrading system development starts with a specification: a printable template, ambiguity tests, failure behavior, acceptance tests and stage gates to live.
- Custom trading software: when to build and what to askNovember 6, 2026 · 12 min readCustom trading software: build, buy or extend open source, which scopes need a license, what stays with you when you outsource and what to ask a vendor.
Do trading bots work, and are they legal?
What regulators and research say, where the rules differ by country and how to recognize a scam.
- Do trading bots work? What regulators and research showNovember 6, 2026 · 13 min readDo trading bots work? As software, usually. As a way to make money, no audited proof was found. What SEBI, ESMA, the BIS and retail trading studies measured.
- Are trading bots legal? US, UK, EU, India, Canada, TurkeyNovember 6, 2026 · 13 min readAre trading bots legal? A table dated 5 November 2026 separating firm duties from private traders in the US, UK, EU, India, Canada and Turkey. Not legal advice.
- Trading bot scams: regulator warnings and how to checkNovember 6, 2026 · 13 min readTrading bot scams from regulators' own cases: fake bots, invented balances, frozen withdrawals. Where to check a seller by country, what to do if you were hit.
Rules people automate
How each type works, what it assumes about the market and how it fails. Examples are hypothetical, and none of them is a recommendation.
- Grid trading bot: how grid strategies work and failUpdated November 5, 2026 · 13 min readGrid trading bot mechanics: arithmetic and geometric spacing, fee drag, inventory risk, a tested Python simulation and the failure modes of grid strategies.
- DCA bot explained: safety orders, settings and risksUpdated November 5, 2026 · 12 min readDCA bot explained: scheduled averaging vs safety orders, how size and deviation multipliers drive capital, the martingale link and a DCA bot vs grid bot table.
- Crypto arbitrage bot: how it works and its real limitsUpdated November 5, 2026 · 13 min readHow a crypto arbitrage bot works: cross-exchange, triangular and funding arbitrage, full cost stack, latency and inventory risk, and why visible spreads vanish.
- Arbitrage bot development: what the build really involvesNovember 5, 2026 · 13 min readArbitrage bot development as a build: pre-funded balances, leg risk, partial fills, fee and transfer data, clock sync, rate limits, failed-leg reconciliation.
AI and trading bots
What AI tools can and cannot do around a trading bot, and how to limit what an AI agent may do on an account.
- Do AI trading bots work? What the evidence actually saysNovember 6, 2026 · 12 min readDo AI trading bots work? AI describes how a bot is built, not whether it earns. What benchmarks and regulators say, how to read a sales page, what to ask first.
- ChatGPT trading bot: what breaks and how to test the codeNovember 6, 2026 · 13 min readA ChatGPT trading bot can place an order and still fail on lost replies, partial fills and restarts. The faults to check, with a tested Python harness.
- Claude trading bot: coding agents, permissions and testsNovember 6, 2026 · 13 min readA Claude trading bot built with a coding agent starts with what the agent can touch: live keys, deny rules, sandbox limits, skills and tests from the docs.
- LLM trading bot: benchmark evidence and a pre-trade gateNovember 6, 2026 · 13 min readLLM trading bot evidence read from the FINSABER, StockBench, DeepFund and LiveTradeBench papers, plus a pre-trade gate tested on malformed model output.
- Machine learning trading bot development: tested failuresNovember 6, 2026 · 13 min readMachine learning trading bot development: a leakage test on random data, feature parity, rollback, NaN at inference and why RL results can mirror the simulator.
- Crypto trading MCP servers: how they work and key risksNovember 5, 2026 · 13 min readCrypto trading MCP servers explained: what they expose, what Binance, OKX and Bitget document, prompt injection risk and how to limit an AI agent's permissions.
Open-source frameworks and libraries
What open-source trading frameworks and libraries do, according to their own documents and repositories, and what they leave for you to build and run.
- Open source crypto trading bot: licenses, status, checksNovember 6, 2026 · 13 min readOpen source crypto trading bot reference: license, language, last release and paper mode for seven projects, read 5 November 2026, and how to check a repo.
- Freqtrade tutorial: Docker setup, dry run and FreqUINovember 6, 2026 · 13 min readFreqtrade tutorial tested on 5 and 6 November 2026: Docker Compose commands, a dry-run config with no keys, safe FreqUI access and the failures you meet.
- Hummingbot tutorial: Client, Condor, API, vs FreqtradeNovember 6, 2026 · 13 min readHummingbot tutorial and map, tested 6 November 2026: Client, Condor, API, Gateway, the unmaintained Dashboard, a paper order and facts against Freqtrade.
- Nautilus Trader Python guide: versions, adapters, limitsNovember 6, 2026 · 13 min readNautilus Trader Python guide: why pip installs v1 while the docs describe v2, a tested one-order backtest, adapter status, no UI and what you must build.
- CCXT Pro: free since 1.95, WebSocket streams testedNovember 6, 2026 · 13 min readCCXT Pro is free inside the MIT-licensed ccxt package since 1.95. Tested watch_order_book, watch_trades and reconnects in Python, read 6 November 2026.
Ready-made bots and platforms
What bot platforms and bots sold as products offer, and what you actually own when you buy one.
- 3Commas alternative: what each platform's page saysNovember 6, 2026 · 13 min read3Commas alternatives by kind, not rank: what each vendor's own page said on 6 November 2026 about keys, withdrawals and regions, plus a checklist for switching.
- Trading bot for sale: what you actually buy and ownNovember 6, 2026 · 13 min readTrading bot for sale: what each form of sale gives you, from compiled files to source code, whether you can move or resell it and what to ask before paying.
Cost and choosing a developer
What drives the price of a custom bot and of a subscription, and what to check before you hire a person or a company. Fincode sells this kind of work, so these guides are not neutral on that point.
- How much does a trading bot cost? Drivers and price rangesUpdated November 5, 2026 · 13 min readThe drivers of custom trading bot cost, running costs, pricing models and red flags, with published seller figures labeled as such.
- Crypto trading bot price: subscriptions, fees, build costUpdated November 5, 2026 · 11 min readCrypto trading bot price models as of November 2026: subscriptions, credit plans, volume and profit fees, self-hosted licenses and a total-cost formula.
- Crypto trading bot developer: what to check before hiringNovember 5, 2026 · 11 min readHiring a crypto trading bot developer? What to specify, what to demand, how keys are handled, how to write acceptance tests and when to walk away.
- Hire a quant developer: roles, test tasks, ownershipNovember 6, 2026 · 13 min readHow to hire a quant developer: which role you need, what to specify, a paid test task with hidden checks, who owns the code, and the red flags.
- Crypto bot development company: how to evaluate a vendorNovember 5, 2026 · 11 min readHow to evaluate a crypto bot development company: evidence to ask for, hosted or client-run deployment, contract terms, exchange API terms and bots to decline.
When a bot misbehaves
A fault-finding order for a bot that stopped trading or trades wrongly.
Subscribe, use open source or commission?
Three common routes and what each one leaves to you. This is a summary of the guide named in each row, not a ranking or a recommendation, and none of these routes makes a strategy profitable.
| Route | What it is | What stays with you | Read |
|---|---|---|---|
| Subscribe to a bot platform | A hosted product, priced by subscription, credits, trading volume or profit commission | Exchange fees, slippage, key permissions and supervision | Crypto trading bot price |
| Run open-source software | Code you can read, run and change under its license | Hosting, keys, updates and results | Open source crypto trading bots |
| Commission custom software | Software built from rules you have written, priced by scope rather than a price list | The rules and decisions, running costs, and maintenance when exchange APIs change | Trading bot cost |
Calculators
- Crypto position size calculator: risk, stop and lot stepEnter your balance, risk per trade, entry and stop prices, fee rate and lot step to see the fractional quantity that keeps a loss at the stop within your chosen amount. Free, no sign-up required.
- Crypto trading fee calculator: maker, taker and slippageEnter the trade size, your maker and taker rates, the share of maker fills, slippage and round trips per month to see what trading costs per round trip and per month. Free, no sign-up required.
- Free crypto profit calculator: P&L, fees and break-evenEnter the buy price, the sell price, the quantity and both fees to see the net profit or loss, the return on cost and the break-even sell price of a spot trade. Free, no sign-up required.
- Crypto average cost calculator: average down with feesEnter up to five buys with price and quantity, plus your fee rate, to see the average cost per unit, the total quantity and, with a current price, the profit or loss. Free, no sign-up required.
- Crypto DCA calculator: average cost from your own pricesEnter the amount per purchase, up to twelve past or hypothetical prices and the fee rate to see the average cost, the total quantity and how the average compares with the plain average of the prices. Free, no sign-up required.
- Grid bot calculator: spacing, fees and net spread per gridEnter the price range, number of grids, spacing type, investment and fee rate to see the mechanics of each grid: spacing, quantity, fees and net spread at the lowest and highest level. Free, no sign-up required.
- Crypto arbitrage calculator: net spread after all costsEnter the prices you see, the taker fees, the withdrawal fee, slippage and size, or three rates for a triangle, to see what is left after costs and the spread needed to break even. No live prices and no scanner. Free, no sign-up required.
- How to buy a trading bot: subscribe or build calculatorHow to buy a trading bot starts with what you are paying for. Type your own subscription and build figures to see the cumulative cost of each by month and the month the build catches up, if it does. Free, no sign-up required.
- TWAP calculator: split an order into timed child ordersEnter the total quantity, the duration and an interval or a slice count to get a schedule of child orders with lot step rounding, a minimum notional check and requests per window. Nothing is assumed about fills or price. Free, no sign-up required.
From Fincode
- Custom crypto trading bot development, built from your written rulesServiceCustom trading bots built from rules you have written, with risk limits, an emergency stop, logs and alerts. Backtested and paper traded before go-live.
- Trading bot code review, audit and repairServiceTrading bot audit: a review of existing code for API key handling, errors, rate limits, order state and logging, in a written report ranked by priority.
- Trading bot maintenance through exchange API changesServiceTrading bot not working after an exchange API change? Maintenance keeps your software compatible and tested. No strategy changes and no position monitoring.
- Technical due diligence and product review for fintech projectsServiceTechnical due diligence for fintech projects: product, market, architecture, code and security review, with a prioritized findings report and a roadmap.
- How we work and what you receiveFincodeHow Fincode runs a software project in four stages: what you approve and receive at each one, what we need from you, and how acceptance testing works.
- Order gateway reference: idempotent order placement in PythonFincodeTrading bot architecture example in Python: idempotent orders, safe retries and order reconciliation on a local mock exchange. Teaching code, real test output.
Frequently asked questions
Does a trading bot make a strategy profitable?
No. A bot executes written rules reliably and removes manual steps. Whether those rules make or lose money is a separate question, and automated trading can lose money.
What does a custom trading bot cost?
It depends on scope rather than a price list: the number of exchanges and assets, the data the rules need, testing depth, risk features, hosting and monitoring. The trading bot cost guide lists the drivers, including what continues to cost money after launch.
These guides are general information about software and markets, not investment, legal or tax advice. Automated trading can lose money, past or backtested performance does not guarantee future results, and the rules that apply depend on your jurisdiction and your exchange's terms. For readers in Turkey, Article 99/A of Capital Markets Law No. 6362 treats activity by foreign crypto-asset platforms directed at Turkish residents as unauthorized.
Other guides
Send a short project brief
The form or an email is enough. You do not need to share your rules yet.
- The exchange or platform you use
- What the software should do
- Your target timeline
- Project briefs and business inquiriesinfo@fincodelabs.com
- WhatsApp (messages only)+90 545 697 41 15 (opens in a new tab)