We productize the problems the field keeps handing us.
Every product here started as a problem an engagement kept running into. That is the sequencing rule: we build what the practice proves is real. One product is live. The others are in development, and we label them that way.
AuditGraph
AuditGraph models the identity estate as a graph and computes consequence from configuration — role assignments, scope hierarchies, federation, PIM eligibility. Because it reads architecture rather than telemetry, it answers on the tenants that do not run deep logging. Azure is generally available.
Full capability detail, pricing and the trust center live on auditgraph.ai — this page is about the company that builds it.
The same practice, asking the next question.
None of these is available today and no dates are promised. They are listed because a roadmap you can read is more useful than one you cannot — and which ships first is decided by our engagements.
Only AuditGraph exists today. The ordering is deliberate — build the identity substrate first, or every later product rediscovers the same estate and disagrees about it.
AI identity is not a separate product.
An AI agent authenticates, holds role assignments and reaches data — which makes it a non-human identity. We model it as a subtype rather than a separate species, so it inherits every engine that already exists for service principals and managed identities.
A standalone AI identity product would be the easier pitch. It would also mean two answers about the same estate. We would rather ship one graph that is right.
Have a difficult cloud problem? Talk to the people who will solve it.
Describe the environment and the problem. We come back with a practical plan — or an honest answer that someone else is a better fit.