LUTFLOW FOR CTOS & PLATFORM ENGINEERS

Enforcement runs in your process, not your cluster.

CTOs and platform engineers need budget enforcement that doesn't require cluster-level access. The LUT Agent SDK wraps your existing OpenAI/Anthropic/Gemini client and enforces policy in-process — no cluster access, no application-layer rewrite. Sentinel adds monitoring-only, metadata-only telemetry from Kubernetes/GKE, with zero application code changes.

The LUT Agent SDK enforces budget policy where the call is made.

JOIN THE WAITLIST →

"Enforcement runs in your process, not your cluster."

🏗️

Monitoring-only infrastructure telemetry

The Sentinel agent deploys as a Kubernetes sidecar or DaemonSet. Zero application code changes. It collects GPU utilization metadata and reports it downstream — it does not enforce policy itself.

⚡

In-process enforcement, no cluster access

Budget enforcement happens inside your application process via the LUT Agent SDK — raise, warn, callback, or SIGKILL the current process on breach. There is no cluster-level kill path today.

☸️

Kubernetes & GKE native telemetry

Sentinel is designed for containerized AI workloads. Helm charts, DaemonSets, resource limits — all integrated. Works with existing Kubernetes monitoring stacks.

🏭

Factory — in development

Self-hosted BYOC deployment into your own cloud is on the roadmap, not shipped today. The active path for self-hosted cost comparison is the Tier 0 measurement CLI (pip install lutflow).

🔧

LUT Agent SDK

For application-level enforcement: pip install lutflow. Three lines of code to wrap OpenAI, Anthropic, or Gemini clients with budget policy.

📡

Metadata-only observability

GPU utilization and inference throughput metadata are visible in real time from Sentinel. Integrates with Grafana dashboards for enterprise monitoring.

Ready to enforce your AI budget?
In-process budget enforcement · No infrastructure changes
JOIN THE WAITLIST →