Product 01 available to businesses

The Loom

Raw exchange feeds in. Verified, fast, query-ready databases out. The Loom is our proprietary engine for turning market noise into data you can trust from day one.

01how it workstap the stages

Where the proprietary part happens.

Every feed passes through four in-house stages before it becomes a database. Select a stage, or an output, to see what it does.

Fig. 1: The Loomlive feeds → database
EXCHANGEReal-time streamsWebSocket EXCHANGESnapshots, archivesREST APIs YOUR FEEDSVenues you chooseby arrangement PROPRIETARY CORE 01 CAPTURE 02 VERIFY 03 REPAIR 04 WEAVE DATABASEFast, columnar, queryable DATASETSVersioned, reproducible THE TRELLISResearch, ready to start YOUR SYSTEMSWarehouse, models, risk
Overview

From noise to a database, without a setup project.

Feeds come in raw and unordered. Inside the core, every message is captured, checked, repaired and woven into structure. What leaves is clean data your teams can query immediately.

Tap a stage or an output

02two productsone fabric

The Loom makes the data. The Trellis puts it to work.

Businesses can take either. We run both ourselves.

standalone data pipeline

The Loom

For firms that need market data they can trust: captured live, verified, repaired and delivered straight into fast, efficient databases. Your venues, your instruments, delivered as a database, data lake or files, with a coverage report on every dataset. Now welcoming early design partners.

available

research platform

The Trellis

Lets a firm start research on day one instead of after months of setup. The Loom's pipeline flows directly into it, and almost everything can be shaped to the way your team works. For banks, funds, brokers and research teams.

in development

03build yoursa sketch, not an order

Sketch the pipeline you need.

Pick what matters to you. Nothing is sent until you press send in your own mail app.

Markets
Data
Delivered into
Used for
Your pipeline, draft
Send this sketch to Lattice Chiti

04who it's foranyone who can't afford bad data

Built for teams that need to trust the data.


Engineering inside.

The concepts we build on. How we combine them stays in-house.

  • Real-time WebSocket streaming
  • REST API integration
  • Sequence-level verification
  • Automatic gap recovery
  • Any-depth order-book capture
  • Columnar storage
  • Point-in-time queries
  • Look-ahead-free backtesting
  • Data-coverage reporting
  • Versioned, reproducible datasets