Build & AI

Polymarket and Kalshi Trading Bots, Built to Your Strategy

Trading bots for Kalshi, Polymarket and other event exchanges, built to your strategy. Settlement rules, related contracts and hard limits are built in.

Updated

A prediction market bot trades contracts that pay out if an event happens. The price of a contract reads as the market’s probability of that event. That makes the strategy feel simple. The engineering is harder, because these venues settle, charge and restrict access in ways a stock or crypto bot rarely meets.

We build bots for prediction markets to your strategy. Quoting both sides, trading on news or a data feed, or trading the gaps between related contracts. You own the code.

Crypto, sports exchanges and broker APIs are on the algorithmic trading page.

What is different on a prediction market

The rule that settles the contract. Each market names its source and its deadline. A bot that reads only the price misses the cases where the wording decides the outcome. We read the rules of every market type the bot will trade, and the bot refuses a market type it was not built for.

Related contracts should add up. Take the brackets of one number. The prices of all the outcomes of one event should add up to one. When they don’t, a strategy may be looking at a real gap or at stale quotes. The bot needs every book of the event at the same moment to tell which.

Money stays tied up. Capital stays in a position until you sell it or the event resolves, sometimes months later. Sizing has to count that, alongside the risk.

Prices jump. A score, a ruling or a published number can move a contract from 30 to 90 cents in seconds. Resting orders are exposed when that happens. A quoting bot pulls them when the signals you choose fire.

Kalshi and Polymarket in November 2026

Both venues change their APIs often, so every fact below links to the venue’s own page, and the date is the day we read it.

Kalshi. Its API covers market data and trade execution over REST, WebSocket and FIX. Every authenticated request draws on a budget of tokens set by your account’s tier, and most requests cost 10. Kalshi’s order groups cancel orders automatically when a contracts limit is reached within a rolling 15 seconds. We use them as a second limit, behind the bot’s own. Kalshi also runs a market maker program for firms that meet its quoting and volume requirements.

Polymarket. The order book has run on CLOB V2 since 28 April 2026, and trades settle in pUSD, a USDC-backed token on Polygon (changelog). New markets move to Protocol V2 in phases from 23 November 2026 (migration guide). A bot written before that date needs its order signing checked. The international platform lists the United States as close-only, and Polymarket publishes a geoblock check. Our bots call it before they send orders. Polymarket US is a separate exchange under CFTC oversight, with its own API.

All of the above was read on 16 November 2026.

What you get

  • The settlement rules of every market type the bot will trade, read and listed, with the bot held to those types
  • Every outcome of an event in one view, so related contracts are priced at the same moment
  • Quoting, news-driven or data-driven entries and exits, written to your rules
  • Limits per market, per event and in total, counting the capital locked until settlement
  • A kill switch, and resting orders pulled when the signals you choose fire
  • Positions and settlements reconciled against the venue's own records

How it runs

  1. Read the venue

    Who may trade from your account's country, the API terms, the fees and the settlement rules of each market type. Anything that blocks the strategy shows up here, before any code.

  2. Write the rules and test them

    Your strategy as rules, tested on the venue's trade history where its API provides it. Order books recorded under your own account from the first week of the project fill the gaps.

  3. Build the bot

    One view of every outcome per event, orders recorded before they are sent, and limits in a layer the strategy cannot change.

  4. Run it in stages

    Paper trading on live books, then real orders at a size you choose. Size goes up only when you approve the step.

Questions we get asked

Do Kalshi and Polymarket allow trading bots?

Both publish official APIs for placing orders from a program. Kalshi documents REST, WebSocket and FIX access with rate limits per tier. Polymarket publishes order APIs and official SDKs. Each has its own rules on who may trade.

Can you build a Polymarket bot for a trader in the United States?

Not for the international platform. Polymarket's docs list the US as close-only there, so a US user cannot open new positions. Polymarket US is a separate exchange regulated by the CFTC, with its own API. We build for the venue your account is on.

Can you build a market-making bot?

Yes. Quoting both sides of a book is one of the strategies we build, with limits on inventory per market and per event. Kalshi runs a market maker program with its own quoting and volume requirements. Whether to apply is your decision.

What data do you test on?

The venue's trade history where its API provides it, and order books recorded under your own account from the start of the project. A strategy that depends on depth needs weeks of recording before the test is fair.

Who owns the bot?

You do, with the code in your repository from the first commit. We sign an NDA before you share the strategy.

Further reading

Start a conversation

Let's scope the work.

Book a strategy call