PLATFORM ARCHITECTURE
HOW IT WORKS
From raw signal to root cause,
in four steps.
Thirty-plus sources enter the context graph. What comes out is the one thing your team actually needs: the answer, with its blast radius, owner, cost, and the commit that caused it.
BUILT FOR
Context at every
One graph. Three audiences. The same underlying relationships
surfaced at the resolution each role actually needs.
Same graph, three altitudes.
Everyone sees the system, at the resolution they need.
AMNIC RADIX · CONTEXT LAYER
TRUST &SECURITY
Enterprise-grade
from day one.
Your data. Your infrastructure. Your control. Radix deploys inside your environment. Engineering signals and system topology never leave your cloud.
Deployment
VPC-isolated
YOUR CLOUD
Radix runtime
ingests · normalizes · graphs
OUR CLOUD
Control plane only
auth · UI · no signal data
01 · INGESTION
Read-only by default
All ingestion via read-only API. No agents. No SDK. No code changes inside your services.
API · read-only
no agents
02 · DEPLOYMENT
Runs in your cloud
VPC-isolated deployment. Engineering signals and system topology never leave your environment.
VPC · isolated
BYOK ready
03 · ACCESS
RBAC + SSO
Scoped access by team, module, and data surface. SAML, OIDC, and SCIM supported out of the box.
SAML · OIDC
SCIM
04 · AUDIT
SOC 2 Type II
Immutable, exportable audit log. Every action, human or agent, fully traceable.
immutable log
SIEM export
Compliance & attestations
SOC 2 Type II
GDPR
ISO 27001
Frequently Asked Questions
1. What is Amnic Radix?
Radix is a context layer that sits across your entire engineering organization and makes sense of how your systems actually behave in production. It pulls in signals from your cost tools, runtime monitors, deployment pipelines, and AI usage logs, then builds a context graph that maps the real relationships between every service, team, commit, and dollar. Instead of jumping across six dashboards trying to piece a story together, you get one connected source of truth that already knows what broke, who owns the fix, and how much it just cost you.
2. How is Radix different from monitoring tools?
Observability tools are excellent at telling you what is happening inside a single system, like CPU spikes or slow queries or error rates, but none of them can connect a latency spike to the deploy that caused it, the team that owns it, and the cost it just added to your bill. Radix sits on top of your existing stack and correlates signals across every tool you already pay for, so root cause, blast radius, ownership, and cost finally show up together in the same view.
3. Do I need to install agents or change my code to use Radix?
Radix is read-only by default and connects entirely through your existing APIs, which means no agents running on your hosts, no SDKs to wire into your applications, and no engineering tickets to ship before you see value. Your team can keep building while Radix quietly maps everything in the background, and most customers are fully live in under a week.
4. With AI generating more of our code, how will Radix help with visibility into what each change actually affects?
AI coding tools like Cursor ship code at a speed your old review processes were never designed for, and they do it without any understanding of your dependency graph, your SLO thresholds, or the cost envelope of the service they are about to change. Radix treats every change the same way, regardless of whether a human or an agent wrote it, tracing each one through the context graph to surface blast radius, ownership, and cost impact directly inside the pull request before merge. So your engineers get the velocity benefits of AI without the production surprises waiting on the other side of the merge.
5. How does Radix attribute cost across services, teams, and AI token spend?
The context graph traces every dollar back through the deploys, services, and commits responsible for it, so cloud spend gets attributed to a specific team or feature without any manual tagging project. The same logic applies to AI token spend, which gets mapped back to the repositories and teams actually consuming it, so the conversation can finally shift from "who spent this" to "was it worth it." Unit economics like cost per tenant, feature, or transaction roll up automatically from there.
6. Will Radix replace tools we already use?
Radix is built to make your existing stack smarter, not to displace it. You keep your APM, your cloud cost tool, your observability tools, and everything else your team already relies on. What Radix adds is the connective tissue between them, stitching their signals together so the tools finally feel like they are part of the same system rather than separate islands fighting for your attention.
7. Where does my data live in Radix?
Your data stays exactly where it already does, which is inside your own cloud environment. Radix runs in a VPC-isolated deployment that you control, and your engineering signals and system topology never leave your infrastructure. The only thing that lives in our cloud is the control plane, which handles authentication and the user interface, so none of your sensitive system data ever leaves your environment.
8. How is Amnic FinOps different from Amnic Radix?
Amnic FinOps is purpose-built for managing cloud costs at scale and preventing spend drift before it compounds, which makes it the right entry point for teams focused primarily on the cost side of the house. Amnic Radix is the broader engineering intelligence platform built on the context graph, connecting cost with runtime, deploys, AI usage, and ownership so every change can be traced to its full downstream impact. Many customers run both, with FinOps handling the cost layer and Radix providing the wider context, and the two are designed to work together as one connected system.
9. How is Amnic Radix priced?
Radix pricing is tailored to your organization rather than slotted into a fixed tier list, because the value scales naturally with two things: the size of your engineering team and the number of tools we connect into your context graph. Pricing is built around your actual setup, so you are not paying for capacity you do not need or running into hidden ceilings as your stack grows.
10. Can I try Radix before committing?
Absolutely, and we strongly recommend it. The easiest way to start is with a 14-day audit of your engineering systems, which is fully read-only and comes with no commitment, so you get to see Radix run against your actual environment without signing anything up front.
