DataByte
Documentation

Docs are opening in stages.

Public product documentation is being migrated from customer-only portals to our website. Rather than ship a half-indexed dump, we're opening documentation track-by-track as each one is reviewed. Here's what's coming and how to get early access today.

What's being written

Three documentation tracks.

Each track is curated by the engineering team that owns the corresponding modules. You can request early access to the technical preview while the public build stabilises.

Getting started

Your first pipeline end-to-end: ingest, transform, deploy a model, expose it as an API, and monitor it.

  • · Environment setup and access
  • · Connect your first source
  • · Build a Transformer pipeline
  • · Publish a Data Insider API
  • · Catalog, lineage, and RBAC essentials
Module references

One reference per module: capabilities, configuration, operational runbooks, and troubleshooting.

  • · Data Ingester: modes, connectors, and schema handling
  • · Transformer Module: canvas and Spark-native patterns
  • · ML Studio and Forecaster: pipelines and deployment
  • · Sherlock: RCA trees and auto-remediation
  • · Platform Admin: RBAC, SSO, and audit
Architecture & operations

Reference deployments, data-plane/control-plane split, and runbook patterns for production.

  • · Deployment topologies (cloud, on-prem, hybrid)
  • · Data residency and key management
  • · Observability and SLA design
  • · Backup, DR, and upgrade procedures
  • · Security posture and compliance mapping
Request access

Need docs today?

Customers under evaluation get read access to the technical preview documentation portal, covering every module, deployment guides, and the API reference. Email us and we'll send credentials within one business day.

What you'll get today
  • A read-only account on our documentation portal with every module reference and architecture guide available.
  • An architecture review call with a DataByte engineer covering your specific stack.
  • Sample pipelines and reference implementations you can run against non-production data.

Prefer a live walkthrough?

A DataByte engineer will walk you through the docs that matter for your use case on a 30-minute call.