Ask Leasyd AI
An AI SRE that investigates your data and answers with the evidence: the numbers, the error messages and the trace behind them.
The AI SRE observability stack
Logs, traces, metrics and uptime in one place, with an AI SRE on call. Predictable pricing per record. Real people answering support.
7 days free, no card needed. Want to see it first? Book a demo: we'll walk you through Leasyd, on your own systems if you like. Moving from another tool? We'll help you point your OpenTelemetry setup at Leasyd: talk to us.
Overview
OpenTelemetryPromQLSQLKubernetesAWS LambdaMCPJavaNode.jsPythonRubyGo.NETOpenTelemetryPromQLSQLKubernetesAWS LambdaMCPJavaNode.jsPythonRubyGo.NET
Works with the tools you already use
Keep everything, for less
One price per record, no plans, seats or minimums. Keep every trace, or drop the noise as it arrives and pay only the small ingest price for it.
Explore pricingKeep up to2.7x more datawith the same budget
or save up to63%with drop rules
1 TBtraces per month
1 TBlogs per month
1 TBmetrics per month
Leasyd, all kept
approx. $1,165 per month
Leasyd + drop rules
approx. $430 per month
An estimate only, at list prices. Assumes an average record of 1 kB (1 billion spans, log records or metric data points per TB), kept 30 days. "Drop rules" keeps 30% of each and drops the rest as it arrives (charged for ingest only). Search, dashboards, alerts and every teammate included.
AI SRE
Explore AI SRE✳What's broken right now?
＋leasyd · 9 tools
list_servicessearch_logsget_tracequery_promql
Robust MCP server
Connect Claude, Cursor or VS Code to your logs, metrics, traces, checks and alerts with Leasyd's read-only MCP server.
Live insights
What changed in the last 30 minutes against the period before, with one click to investigate.
Read-only by designNever trains on your dataAnswers link to traces
Tracing
Explore tracing
Every span of every request
Follow a request through every service in one waterfall, with its logs, attributes and exceptions side by side.
Outliers at a glance
A heat map of span durations: slow and failing requests stand out in red, so you start where it hurts.
SQL over raw spans
When you need an exact answer: read-only SQL over your logs, spans and metrics.
OpenTelemetry nativeNo sampling required30 days of history
Uptime monitoring
Explore uptime monitoring
HTTP and browser checks
Check APIs and pages every minute, from single requests to multi-step logins in a real Chrome browser.
SLOs and error budgets
Track objectives such as 99.9% under 300 ms, and see how much of the budget is left.
Alerts where you work
Email, Slack or webhook when a check fails or a query crosses a threshold. Maintenance windows keep planned work quiet.
1 to 60 minute checksEncrypted secretsScreenshots of every step
Log management
Explore log management
Search every log line
30 days of logs, searchable in seconds by service, severity, text or any attribute, one click from the request that wrote them.
Debug logskeep 0%
severity_number < 9
Health checkskeep 0%
name = GET /healthz
Fast, successful spanskeep 10%
duration_ns < 50000000
Preview: 62% of the last 24 hours matched
Drop rules
Discard debug logs and health checks as they arrive: dropped data costs the ingest price only, never storage.
Groups and patternsJump to the tracePromQL over log counts
Metrics & dashboards
Explore dashboards
Dashboards out of the box
Built-in dashboards for services, JVM, Node.js, Python, Ruby, hosts and AWS Lambda; build your own panel by panel.
Query builder and PromQL
Build a query with clicks or write PromQL, with percentiles from your histograms.
Prometheus-compatible APIAWS CloudWatch metricsShareable dashboards
Find the cause, not just the symptom
Start a 7-day free trial: every feature, up to 1 GB of data a day, no card needed.
© 2026 Leasyd Ltd · Registered in England and Wales, company number 172858754-6 Greatorex Street, London E1 5NF