One platform,
four surfaces.

XTrace is one core platform, configured to your KPIs and outcomes. Each surface is built for the people who use it, and all four share the same data.

Surfaces

Built for the people who use it.

Results at a glance, a place to run the program, and help inside the tools people already use.

CFO, CIO and CTO offices

Spend & ROI

Portfolio spend, a spend forecast, cost per outcome and budget against actual. Used now and then, and clear at a glance.

FinOps, platform and AI enablement teams

Policies & recommendations

Where the work gets done: tagging and allocation rules, budgets, policies, anomaly alerts, routing configuration and recommendations to approve. Its adoption and coaching view shows where teams get stuck, what good practice looks like and which standards to roll out.

Everyone who uses AI tools

The XTrace plugin

Works with the AI tools and coding agents your teams already use. It suggests the right model in the moment, trims prompts and context, keeps budgets in view and reuses what the team already knows. It enforces company policies and stops known mistakes inside the session, and each new fix XTrace finds reaches it automatically. Nearly invisible when everything is fine.

Runs in the background

AI gateway

Sits in the request path for agents and apps and runs the routing, caps and caching your team has set. It analyzes traffic and recommends changes, and a person approves each one in XTrace. It makes no change on its own.

Inside the platform

What each view is for.

A high-level look at the views teams use most. Every one reads the same data.

Spend & ROI

Executive spend overview

Total AI spend at a glance for the CFO, CIO and CTO offices: where it goes, budget against actual, and a forecast from real usage.

From

AI providers, tools and billing

XTrace

One view of spend and budgets

You see

The forecast, and the forecast with suggested optimizations

Spend & ROI

ROI viewer

Spend next to the outcome data your program tracks, so each team’s cost per outcome is plain to see.

From

Spend, and the outcome data your program tracks

XTrace

Matches cost to outcomes, by team

You see

Cost per outcome, against the KPIs you report on

Policies & recommendations

Work quality

Where AI work goes wrong, team by team: failing tools, loops and missing context. Every finding is tracked until the fix is verified.

From

AI sessions, at team level

XTrace

Finds repeat mistakes and policy violations

Result

Fixes the XTrace plugin enforces automatically

Policies & recommendations

Suggested optimizations

Changes that would lower cost or lift quality, such as a model or skill change, with estimated savings where XTrace can calculate them.

From

Usage, spend and work quality

XTrace

Suggests what to change

You decide

Approve the changes you want

Policies & recommendations

Skill optimizer

Shows which shared skills earn their keep, and suggests improvements for the ones that don’t.

From

The skills your teams share

XTrace

Compares how each one performs

You see

Suggested improvements

AI gateway

Router management

Routing, caps and caching for every team, set in one place. The AI gateway runs them on every call and recommends changes; a person approves each one.

From

Every model call

AI gateway

Runs your routing, caps and caching

You decide

Recommended changes wait for your approval

And more

Automatic anomaly alerts, budgets, adoption and coaching, and reports in the app or on a schedule.

How they fit together

One control point for every AI call.

The AI gateway sees every call and recommends changes, and a person approves each one in XTrace. Rules and policies reach the XTrace plugin automatically, and it enforces them inside each AI tool.

  • Request path. Every model call from AI tools, agents and apps passes through the gateway.

  • Recommends. The gateway analyzes traffic and surfaces routing, caching and budget changes. A person approves them in XTrace.

  • Enforces. Rules and policies go to the XTrace plugin, which applies them automatically inside each tool.

Where each surface sits. Gateway recommendations go up to a person; rules and policies go down to the plugin in each tool, automatically.

See the four surfaces with your numbers.

We’ll walk you through each surface, set up around the KPIs you already report on.