Aperiodic
Metrics
MarketsLive
DocsPricing
Get Started

Team

The data systems hedge funds have, open to every quant

We started as a mid-frequency trading desk and built the kind of data pipeline funds build for themselves. Aperiodic is that pipeline, opened to anyone who does research.
Talk to usHow it’s built

Shown: traded volume summed over each venue’s perpetuals, per minute, 18:00–01:00 UTC on 10 Oct 2025, each venue on its own square-root scale; the busiest minute, 21:13 UTC at $5.7B across the three venues, is orange · built with ohlcv 1m

Where we come from

We built it for our own desk first

Aperiodic started on our own systematic trading desk. We began in macro, moved to crypto, and needed market data we could trust in a backtest.

Over the years we tried nearly every data provider in crypto. The raw tick archives had everything, plus months of engineering between you and your first regression. With the polished vendors, you met a salesperson before you met the data. So we built our own pipeline, first for our desk, then for partner institutions.

In March 2025 we opened it to everyone: the same data we built for ourselves, with a free preview of every dataset that downloads with one command and no key.

Mar 2025
Aperiodic opens to everyone
Sep 2026
Raw trades and quotes
Oct 2026
Live data
Nov 2026
An MCP server for AI agents

7 years

of market history, back to Sep 2019.

OKXBinanceHyperliquid
2020202120222023202420252026
Shown: the months of history we serve on each venue (OKX from Sep 2019, Binance from Jan 2020, Hyperliquid from Jan 2025); symbols listed later start later. Orange is this month.

Why it’s hard

Recording a market without gaps is most of the work

Anyone can connect to an exchange in an afternoon. Keeping every message, once and in order, through every dropped connection, is the actual job.
Connection AConnection BWhat we keep
Illustrative: two connections carry the same feed and each drops once (orange). The record keeps whichever copy arrived first, so it has no gap.
  1. 01

    Connections drop

    Exchanges close every websocket within a day. We stream each feed on two machines at once and keep the first copy of every message.

  2. 02

    Messages arrive twice, late or out of order

    We spot duplicates without parsing each message, where the exchange makes that possible, and put stragglers back in order inside a short window.

  3. 03

    Every exchange names things its own way

    Atlas, our open-source security master, matches the same asset across a dozen exchanges, every day.

  4. 04

    A bar must only hold what you could have seen

    Our pipeline closes each window on the time the data reached us, not the time the exchange printed on it.

  5. 05

    Machines crash

    The engine saves its state every 10 seconds and never publishes half a window. After a crash it picks up from the last save.

  6. 06

    Definitions change

    A changed metric ships as a new version beside the old one, and the old one keeps computing exactly what it did.

See the whole pipeline

The team

One team, from the exchange socket to your dataframe

Aperiodic is made by Unravel Finance, which also builds Unravel, a research platform for crypto factors. Our own factor research runs on this data too, which is the best reason we know to keep it working.

Capture

200+ producers keep Binance, OKX and Hyperliquid connected around the clock, two on every feed.

Compute

A streaming engine we wrote in Rust turns every message into metrics within 50 ms of a window closing.

Research

We define each metric and explain it, on real data, in the metrics library and the guide.

Delivery

REST, the live feed, the Python SDK, the CLI and an MCP server, plus people who answer when you write.

Leadership

Run by its founders

Mark and Daniel run the company as co-CEOs, and both are happy to talk about your data.
Portrait of Mark Aron Szulyovszky

Mark Aron Szulyovszky

Co-founder & co-CEO

Mark grew his last company, Drops, to $12m in annual recurring revenue before Kahoot bought it in 2020. After the sale he started a family office to run systematic strategies, and at Aperiodic he works on every part of the pipeline, from the first byte in to the last byte out.

  • Book a call
  • LinkedIn
  • X
Portrait of Daniel Szemerey

Daniel Szemerey

Co-founder & co-CEO

Daniel came to markets from computation, with an MSc in Architectural Computation from UCL. He founded a supply-chain startup, ran a GE Healthcare-backed accelerator for more than 150 medical AI companies, and co-founded Unravel with Mark.

  • Book a call
  • LinkedIn
  • X

If you need help with data, talk to us

Tell us what you’re researching and we’ll point you to the data that answers it.

Book a callBrowse the metrics

Shown: minute lows of the 465 Binance perpetuals, 20:40–22:00 UTC on 10 Oct 2025, with BTC’s path traced · built with ohlcv 1m

Aperiodic

Built from hundreds of terabytes of raw data and distilled into point-in-time metrics you can pull as parquet files. Plus the raw trades, quotes and derivatives data behind them.

Use Cases
  • Improve your backtest
  • Find new alpha
  • Execute smarter
Indices
  • All indices
  • Aperiodic Cost Index
  • Aperiodic Depth Index
  • Aperiodic Impact Index
Metrics
  • Metrics library
  • Data Catalog
  • Live markets dashboard
  • Global Liquidity
  • Unusual activity
  • Order Flow
  • L1 & L2
  • Derivatives
  • RWA
  • Raw Data
Developers
  • API Docs
  • MCP Server
  • Notebooks
  • For AI Agents
  • Atlas Security Master
  • Pre-flight
  • Changelog
  • Roadmap
  • llms.txt
Learn
  • Replays
  • Microstructure Guide
  • Architecture
  • Research Blog
  • FAQ
Compare
  • All vendors
  • vs. Tardis
  • vs. CoinGlass
  • vs. Kaiko
  • vs. Amberdata
Company
  • Team
  • Pricing
  • Contact
  • Book a call
  • Terms of Service
  • Privacy Policy
  • Licensing
  • SLA

© Copyright 2026 Aperiodic. All Rights Reserved.

Aperiodic Limited · Registered in Ireland · Company No. 815273 ·
136 Capel Street, Dublin D01 T2C9, Ireland