DB
Back homeIndustry

Crypto & web3

Go from raw blocks to production dashboards without stitching together a queue, a cache, and three databases.

The problem

Why this is hard today

Web3 data teams inherit the hardest ingestion problem and the strictest correctness bar at once: decode raw chain data, survive reorgs, reconstruct entity state, and never return an approximate holder count.

The usual answer — an indexer plus a warehouse plus a cache — drifts out of sync and answers the same question two ways.

A single engine that decodes in-database, reconstructs state temporally, and keeps exact counts turns the pipeline into a schema.

Where NYXDB fits

Use-case journeys

On-chain analytics & decode pipelines

Decode raw blocks into typed relations with continuous transforms and query exact state.

Blockchain & on-chain analytics

DEX & market data

Live pool/price state as keyed tables with streaming reads over swaps.

Market data & trading analytics

On-chain risk & compliance

Read-your-writes address state and exact counts for screening and scoring.

Fraud & risk

Architecture

How NYXDB fits crypto & web3

Raw events append; transforms decode them into typed state; attribute tables reconstruct wallet and contract state as of any block. This is the chainflow shape.

  1. 01

    Ingest raw

    Blocks and logs append, durably journaled.

  2. 02

    Decode

    CREATE TRANSFORM … INTO typed state tables, in-database.

  3. 03

    Reconstruct

    Attribute tables + AS OF give state as of any block.

  4. 04

    Serve

    Exact holder counts and balances power dashboards.

Real SQL

Representative query

Top holders as of a block
SELECT chain_id, name, holders
FROM token_state
FOR SYSTEM_TIME AS OF 42;

Every statement follows the engine’s own test SQL shapes. See the SQL reference for full syntax.

Capabilities

What you get

In-database decode

Transforms replace the external indexer.

Temporal state

Reconstruct as of any block.

Exact counts

Holder counts that are never wrong.

Multi-chain

One schema across chains.

Proof

Measured where it counts

exact

holder counts (keyed collapse)

ADR-078

AS OF

state per block

ADR-071

multi-chainTODO-verify

decode throughput class

placeholder — not yet substantiated

▲Figures marked TODO-verify are placeholders pending a published, reproducible benchmark; substantiated numbers cite their source.

Learn more

Related documentation

Attribute tables

Temporal reconstruction.

Transforms

Decode-to-state pipelines.

SELECT … AS OF

Point-in-time reads.

From blocks to dashboards, one engine

Explore on-chain analytics or run a node.

Get started Read the docs