About SieveCore
Why We Built SieveCore
Production teams should not need redeployments to change log visibility. We're building the governance layer that makes observability behavior controllable at runtime.
Our Mission
Make observability behavior governable at runtime.
Production logging has been treated as a deployment-time concern for too long. We believe it should be a runtime governance concern — with scoped policies, scheduled windows, automatic rollback, and full revision history. Just as feature flags decoupled feature releases from deployments, SieveCore decouples log behavior from the software delivery lifecycle.
The Problem
The Problem We Solve
Production teams should not need a deployment to change log behavior. This is the story every production engineer knows.
The Old Way — Every Incident
Need visibility
An incident occurs. You need more log data to investigate.
Code change
Write a change to enable DEBUG logging. Open a pull request.
Deploy
Wait for CI to build. Push through the deployment pipeline.
Restart
Service restarts. Lose in-flight requests and context.
Investigate
Finally get the logs you needed 30 minutes ago.
Repeat
Redeploy again to disable DEBUG. Hope nothing breaks.
The SieveCore Way
Select a scoped target
Choose the namespace, application, or cluster that needs different log behavior.
Create or edit a policy
Define a default policy or a runtime policy with a time window. No code change. No PR. No CI.
Inspect the effective policy
See which policy is active and why — across the full hierarchy.
Automatic rollback
Runtime policies expire and roll back automatically. Revision history records every change.
Governance at runtime. Not at deploy time.
Ready to stop redeploying for logs?
See how SieveCore gives your team runtime control over production logging — without touching a deployment pipeline.
How We Build
Engineering Principles
Every decision we make is guided by these principles.
Governance-first
Every change is scoped, reversible, and recorded. Nothing happens without a policy. Nothing persists without explicit intent.
Transparency
The Effective Policy Inspector shows why a policy is active. Revision history shows what changed. No hidden behavior.
Runtime control
Changes take effect at runtime. No redeployments. No restarts. The policy engine operates independently of the deployment lifecycle.
Customer-controlled
SieveCore runs inside your Kubernetes cluster. Your data, policies, and audit trails stay in your infrastructure.
Complementary architecture
SieveCore works alongside Datadog, Splunk, Grafana, and major observability platforms. It controls what gets logged — they handle the rest.
Built for operators
Designed for platform engineers, SREs, and DevOps teams who manage production logging at scale. No ceremony. No friction.
Deployment Model
Customer-Controlled Environments
SieveCore is built for organizations that keep production infrastructure under their own control.
Customer-controlled deployment
SieveCore runs inside your Kubernetes cluster. The control plane lives in your infrastructure, not in ours.
Your data stays with you
Policy configurations, revision history, and audit trails stay inside your environment. No external data exfiltration.
Governance by default
Every policy has a time window. Every change rolls back automatically. No accidental permanent modifications.
Works alongside your stack
SieveCore complements Datadog, Splunk, Grafana, and major observability platforms. It controls what gets logged — they handle the rest.
Questions
Frequently Asked Questions
See SieveCore in action.
We're working with early engineering teams to shape Runtime Observability Governance. If changing log behavior through deployments is painful, we'd love to show you the product.