2026.81 August 2026
- Copilot now authors and validates rollback plans alongside forward runbooks
- Wave planner accounts for cross-region replication lag when sequencing
- Audit ledger export to Splunk and Microsoft Sentinel
Resources
Everything we know, without a registration wall
Reference architectures, regulatory mappings, operating-cost research and the templates we use with customers. Download what is useful; talk to us if it is.
Technical brief
The autonomous operations reference architecture
How a 40,000-node estate runs governed change with a nine-person platform team. Deployment topology, policy model and the failure modes we designed around.
Report
Quantifying the cost of manual infrastructure operations
Effort and cost baselines drawn from 40 enterprise estates, broken down by platform, region and change class, with the methodology included.
Whitepaper
Mapping DORA technical requirements to infrastructure controls
Article-by-article mapping from the regulation to testable technical controls, with evidence templates for each.
Guide
Designing a policy model for unattended remediation
How to decide which failure modes may be resolved without a human, and how to bound them so the decision stays defensible.
Webinar
Patch orchestration at availability-group scale
A live walkthrough with Northwind Financial’s database engineering team on compressing a quarterly cycle from eleven weeks to six days.
Template
Proof of concept scoping template
The scoping document we use to define a two-week proof of concept, including the measurement plan that determines whether it succeeded.
Status
Platform status
Live component status and rolling 90-day availability. Incident history, post-incident reviews and the RSS feed are published without authentication.
All systems operational
No incidents affecting customer environments in the last 90 days. The most recent post-incident review was published on 12 April 2026.
Component statusOperational
- Control plane · EU99.99%
- Control plane · US99.99%
- Control plane · APAC100%
- API99.98%
- Copilot inference99.95%
- Connector registry100%
Rolling 90-day availability
Changelog
What shipped, monthly
Infrapilot ships on a monthly release train. Customers on self-hosted deployments receive the same release with a 30-day soak window.
2026.73 July 2026
- OpenTofu support alongside Terraform for generated infrastructure code
- Policy simulation — evaluate a policy change against the last 90 days of runs
- Kubernetes add-on lifecycle management for EKS, AKS, GKE and OpenShift
2026.65 June 2026
- SAP HANA and Sybase ASE added to the managed database set
- Risk model retraining moved to nightly, per tenant
- Approval engine supports delegated and time-bounded pre-authorisation
Support
How to reach an engineer
Enterprise and Sovereign customers have a 24×7 channel with a 30-minute P1 response and a named customer architect who knows your deployment.
Support portal
Raise and track cases, view your entitlement, and reach your named architect.
Documentation
Guides, references, runbook library and migration playbooks, versioned per release.
Community
A moderated forum where customer platform teams share workflows and policy patterns.
Emergency line
A direct number for active P1 incidents, staffed by engineers rather than a triage desk.
Get started
See it run against your own estate
A proof of concept takes two weeks. We deploy inside your network, discover a scoped part of your estate, and run a real change end to end — with your team holding the approvals.