- support@haventech.guru
- Mon - Fri: 6.30 am - 5.00 pm
We are creative, ambitious and ready for challenges! Hire Us
AI Labs / Synapse Alert
AI Labs · Synapse AlertSee your network.
Know the moment it matters.
Synapse Alert is a self-hosted monitoring platform that shows your whole estate as a live, animated map — and a monitoring engine smart enough to page only when something is really wrong.
Talk to us: 312-432-4480Email usBranch firewall — no response 3 checksDOWN
~78 sfrom failure to notification
82%fewer phone alerts — same coverage
16,600checks per hour across 390 devices
69×faster status pages after tuning
The map is the monitor
Your network as a living diagram, not a list of tiles
Every router, firewall, server and circuit on one canvas. Links pulse at a speed set by their measured latency, turn grey when one end is down and dashed when both are.
Map studio
Drag-and-drop editor with 51+ device types, shapes, labels, images, 8 visual themes, undo/redo and auto-save — rendered on the graphics card for smooth motion.
Deep checks
Ping, TCP port, DNS, HTTP/HTTPS with keyword and JSON checks, certificate expiry, heartbeat 'dead-man' checks, and Windows and Linux service checks.
Status pages
Branded public status pages with incident posts and uptime badges — so customers know before they call.
How it decides
The smarts behind every page you receive
# a device stops answering
check critical gear every 20 s · workstations every 5 min
confirm 3 consecutive failures (per-device override)
topology upstream router down? → suppress everything behind it
window in maintenance / quiet hours? → record, don't page
route critical → phone + Teams · warning → email
recover 3 good checks in a row before 'all clear'
result one accurate alert naming the device that actually failed
- Status judged on the last five results: up, degraded, down or unknown
- Cooldowns, re-send intervals and burst digests — 169 weekend alerts became 2 emails without delaying the first page
- Recurring weekly windows, even across midnight: suppress during, or alert only during
- Silent monitoring: recorded in the app, nobody woken up
- Maintenance schedule published by API so other tools honor the same windows
- It proves its own health — an external watchdog so it can never fail silently
Every channel your team lives in
Real-time alerting, routed by severity
Microsoft Teams
Critical, warning, info and recovery routed to the right channel.
Email & SMS
Readable on a lock screen or a watch: state first, then device, address and port.
Slack, Discord, Telegram & PagerDuty
Plus webhooks — with retries and backoff so no alert is lost.
AI-ready
An AI integration lets assistants check status, run diagnostics, open maintenance windows and acknowledge alerts.
Measured in production
Results, not promises
- Detection-to-notification cut from ~150 s to ~78 s
- Critical gear checked twice as often while total load fell 28%
- Phone alerts per day: 34.8 → 6.4
- Zero failed alert deliveries over 7 days
- Monitoring history storage down 77%
- Deploy downtime: ~8 minutes → ~30 seconds
- Full restore verified: 37 tables, millions of records, zero errors
- Two-factor login, 4-level access control, 365-day audit trail
More from AI Labs
Explore the family
Synapse Alert also feeds NetCentral: its maintenance windows tell the operations wall what's planned, so planned work shows amber — not red.
Let’s talk
Want a network map that tells you the truth?
Call or email and a Haventech engineer will walk you through what AI can do for your business. Mon – Fri, 6:30 am – 5:00 pm.