Services
Platform
Industries
EngagementsPricing
Resources
Company
Book a discovery call (470) 784-4045

Support & SLAs

Support with the response times written down

Every Perflens system is covered by a Perflens Care tier with a published Severity 1 response, a clear route to an engineer and replacement hardware that ships before the faulty unit comes back.

A large bearing ring wrapped with orange bands like a lifebuoy, with a coil of rope.
FIG. 48Someone throws you a line — a lifebuoy made from a bearing ring.

Contact support

How to reach the Care desk

Email

support@perflens.com for Severity 2 and 3 issues, questions, configuration changes and report requests. Tell us the site, the station or asset, and what you're seeing.

Phone

(470) 784-4045, Monday–Friday, 8:00 a.m.–6:00 p.m. Central Time. For Severity 1, call rather than email, so the issue reaches an engineer without waiting in a queue.

After hours

Care Dedicated customers have a 24/7 critical line — the same number routes to the on-call engineer outside business hours. On Care Essentials and Care Verified, response times run in business hours.

Service levels

Perflens Care service levels

What each tier covers, and how it fits with your system, is set out on the Perflens Care page. These are the commitments behind it.

Uptime target for on-device inference: 99.9% per month, measured on the Gatepoint. Prices for each tier are on the pricing page.
CommitmentCare EssentialsCare VerifiedCare Dedicated
Severity 1 response4 business hours1 hour30 minutes, 24/7
Alert verification by a Perflens analyst—Within 4 hours, business daysWithin 1 hour, 24/7
On-site visit (contiguous US)ScheduledWithin 2 business daysNext business day
Advance hardware replacement3 business daysNext business dayNext business day plus on-site spares kit
Model retrainingTwice a yearQuarterlyContinuous, as needed
Drift monitoring and monthly health report✓✓✓
Named reliability engineer——✓
After-hours critical line——✓

Severity

Severity definitions

SeverityDefinitionExamplesHow to report
Severity 1A Perflens system is stopping production or has stopped detectingReject gate held in its fail state; Gatepoint offline; a camera returning no image; no data from a critical assetPhone
Severity 2Detection or monitoring is degradedFalse rejects climbing; a sensor node dropping out; alerts arriving latePhone or email
Severity 3A question, a configuration change or a reportAdding a user; changing an alert route; requesting a custom reportEmail

Escalation

The escalation path

  1. Care desk analyst

    First response

    Logs the issue, confirms the severity with you and starts diagnosis using the device health data already in Perflens Console.

  2. Support engineer

    Remote diagnosis

    Connects through the remote-support link if you have enabled it, reviews logs, images and signals, and fixes what can be fixed remotely — a configuration change, an update rollback or a model revert.

  3. Field engineer

    On site

    Travels to your site when the fix needs hands, within the on-site time for your tier, and brings replacement hardware if it is needed.

  4. Leadership

    Severity 1 without a clear fix

    Any Severity 1 without a clear path to resolution goes the same day to our Head of Customer Care and the relevant engineering lead. On Care Dedicated, your named reliability engineer coordinates throughout.

Remote support

Remote first, on site when it matters

Most issues on an edge system are fixed fastest from a keyboard: a configuration change, a rollback to the previous update, a model revert, a camera exposure adjusted after someone changed the lighting. Our support engineers work through an outbound-only, encrypted link from your Gatepoint. There is no inbound port on your firewall, and you can switch the link off from Console at any time.

Every remote session is recorded in the Console audit log with the engineer's name, the time and what changed. If your policy keeps the link off, we support you over a screen share you control and, when needed, on site within your tier's on-site time. The Security & Trust Center describes the connection in detail.

Hardware

Spares and advance replacement

The replacement ships first

When a device fails, its replacement leaves our lab before the faulty unit comes back — within 3 business days on Care Essentials, and next business day on Care Verified and Care Dedicated.

Spares on your shelf

Care Dedicated sites hold a spares kit sized to their stations and assets, and we restock it whenever a part is used.

A swap, not a recommission

A replacement Gatepoint pulls its configuration, models and certificates from Console when it enrolls, so a swap is a cabling job. See hardware.

Updates

Maintenance windows and update policy

Updates should never surprise a production manager. These rules apply to every firmware, Console and model update.

  • Maintenance windows are agreed at handover and recorded in your site documentation
  • No update is installed during production without your written approval
  • Firmware and Console updates are signed and installed to the inactive A/B partition, with automatic rollback if health checks fail
  • Model updates run in shadow mode beside the live model before they take over
  • Release notes and model card changes are sent to you before every update
  • Security fixes are labeled with their severity, so urgent ones can be scheduled ahead of the normal cycle
  • Air-gapped sites receive signed offline update packages

Training

Training for every team that touches the system

Operators

At handover, on your line — reading verdicts and stack lights, handling rejects, clearing simple faults and knowing when to call us.

Maintenance and controls

Cleaning optics, swapping a sensor node, checking I/O and the PLC heartbeat, and restarting the system safely.

Engineers and administrators

Console administration, alert routing, user roles, reports, and how to read a model card.

Refreshers

Sessions when staff change or a new line goes live, run by our Care team on site or online.

FAQ

Questions we hear often

What counts as Severity 1?

Any situation where a Perflens system is stopping production or has stopped detecting. If you are unsure, call and treat it as Severity 1 — we would rather downgrade a ticket than miss a real one.

Can we change Care tiers?

Yes. Tell us and we'll confirm the change and its effective date in writing. Moving to Care Dedicated includes setting up your on-site spares kit and introducing your named reliability engineer.

Are updates included?

Yes. Firmware, Console and security updates, and model retraining at your tier's frequency, are included in every Care tier.

What if our security policy blocks remote access?

Then the link stays off. We support you by phone, over a screen share you control and on site, and updates arrive as signed offline packages. Some fixes take longer that way, so we agree how Severity 1 will be handled before handover.

What is in the monthly health report?

Model confidence and drift against baseline, reject and alert statistics, device health, update status, and any retraining completed or scheduled. Our field note on model drift explains what the drift figures mean.

Not a Perflens customer yet?

Every system we install comes with these commitments. Book a discovery call to find out which Care tier fits your line, assets or building.