AI for IT services and managed service providers
A managed service provider answers for many clients at once, and each client's truth is in the ticketing system, the monitoring, the identity provider and the billing. Memtro connects to them and answers the questions an engineer or an account manager asks, with each fact cited to the ticket, alert or invoice it came from, so a client review is prepared in minutes and an incident is explained from the record.
7-day free trial · card on file, nothing charged until it ends · cancel any time
Tickets, contracts, systems and billing across every client, answered from one memory with sources.
The systems these questions draw on, connected to one memory. Every connector reads by default; the tools that change something are granted by name, per team.
What it looks like today
- 01An incident touches the ticket queue, the monitoring and the on-call tool, and the timeline is reconstructed by hand.
- 02Which clients are over their contracted hours is known at invoicing.
- 03A joiner or leaver request is confirmed done by asking the engineer.
- 04The quarterly client review is assembled from four reports.
- 05The same fix is worked out three times by three engineers, because the first write-up was in a ticket nobody searched.
- 06Every client wants their own report, in their own format, at their own point in the month.
What Memtro does about it
One memory in front of the systems you already pay for, reading by default and acting only where an administrator has granted the tool by name.
- Connect ServiceNow, Jira Service Management, Freshdesk or Zendesk, Datadog, PagerDuty, Microsoft Entra ID, Okta, Microsoft Teams and QuickBooks Online, reading by default and acting where you grant it.
- Ask about an incident and the answer joins the tickets, the alerts and the on-call record into one cited timeline.
- Ask whether a leaver has been offboarded and the answer reads the identity providers rather than asking the engineer.
- Connections are granted to named teams, so each client's systems are visible only to the engineers on that account.
- Schedule jobs: tickets breaching SLA in the next four hours, clients over 80% of contracted hours, accounts still enabled for leavers.
System by system
What Memtro reads in each part of the stack, and what it may change there once the tool has been granted.
Service desks
ServiceNow, Jira Service Management, Freshdesk, Zendesk, Zoho Desk, Help Scout or Front: incidents, requests, changes, SLAs and work notes. Where a desk offers replying, commenting or changing state, each is a granted tool rather than a default.
Monitoring and on-call
Datadog, Grafana, Sentry, PagerDuty, incident.io, CloudWatch and CloudTrail: the alert, the page and the timeline. PagerDuty can acknowledge, resolve, reassign or raise an incident once granted.
Identity
Microsoft Entra ID and Okta: people, groups, applications and who has access to what, so a joiner or leaver is confirmed from the directory rather than from an engineer's word.
Project and change work
Jira, Linear, Asana, monday.com or ClickUp, read beside the ticket queue so a client's projects and their incidents are one picture.
The client's own systems
A PSA, RMM or reporting database over SQL Server, Postgres, MySQL or MongoDB, read-only inside a transaction with a row cap and a timeout, and the SQL shown in the answer.
Billing and chat
QuickBooks Online or Xero for contracted hours and invoices, Microsoft Teams or Slack for where the answers and the scheduled checks land.
There are 144 connectors in all, each with its own tools, its own permissions and a page of its own. Browse every connector →
Who asks what
Service desk engineer
Asks whether this has happened before, at this client or another, and what actually fixed it, instead of working it out again.
Account manager
Opens the client review with the tickets, the incidents, the hours used and what is still open, assembled rather than collated.
Service delivery manager
Watches SLAs before they breach, reopened tickets and leaver requests that were closed while the account is still enabled.
Director
Sees the whole book: hours against contract by client, incident load, and which accounts are quietly consuming more than they pay for.
The service month
- 01
Hour by hour
A job lists the tickets due to breach in the next four hours, by client and engineer, so the queue is managed before the report explains it.
- 02
When something breaks
The timeline is assembled from the alerts, the pages and the tickets into one cited sequence, which is also the first draft of the write-up the client will ask for.
- 03
Joiners and leavers
A weekly job checks closed leaver requests against Entra and Okta, so an account that is still enabled is found by a job rather than by an auditor.
- 04
The client review
Hours used against contract, tickets opened and closed, incidents and what is still outstanding, per client, ready to be read rather than built.
Ask it like this
Plain English, from chat, your editor, a scheduled job or your own agent. In agent mode each question also checks the other systems that know about the same people, companies and references, and the answer ends with a line per system consulted.
- ServiceNow
- Jira Service Management
- Freshdesk
- Zendesk
- Datadog
Every figure in the answer cites the record it came from, and the run ends with a line per system consulted.
- ›Which tickets will breach their SLA in the next four hours, by client?
- ›Has the leaver Sam Lee at BigCo been disabled in Entra and Okta?
- ›Which clients used more than 80% of their contracted hours this month?
- ›Has this problem happened at another client, and what fixed it?
- ›Which clients raised the most tickets this month, and what were they about?
- ›Who at BigCo is in the administrators group in Entra and Okta?
- ›What is still open from last month's client review?
Work that runs itself
A job is a prompt the agent runs on a cron schedule as its owner, with their tools and connections. Say any of these in chat and the job exists; every run's report is kept in a chat you can question.
- schedule:Every hour: tickets due to breach SLA in the next four hours, grouped by client with the engineer assigned
- schedule:Every day 08:00: leaver requests closed in the last week whose accounts are still enabled
- schedule:Every Monday 08:00: hours used against contract by client, with tickets opened and closed
- schedule:Every Monday 07:00: incidents and major changes at each client last week, as the first draft of their report
- schedule:Every day 09:00: tickets reopened in the last 24 hours, with the engineer who closed them
Questions people ask
Can it disable accounts or close tickets?+
Only where you grant it. Zendesk can reply on a ticket and change its status today, Teams can post, and actions on identities and incidents are granted the same way as they arrive: by name, per team, used only when a person asks for exactly that.
Can clients be kept apart?+
Yes. Connections are granted to named teams, tool by tool, and nothing shared is usable until an admin says so. One engineer's view of a client's systems never leaks into another client's answers.
Does it reach our RMM or PSA?+
ServiceNow, Jira Service Management, Freshdesk and Zendesk are covered directly. A PSA or RMM with a reporting database is reachable over SQL Server, Postgres or MySQL.
Can it act on a ticket or an incident?+
Where you grant it. ServiceNow can raise an incident, add a work note and change state; Zendesk, Freshdesk, Front and Zoho Desk can reply and change status; PagerDuty can acknowledge, resolve, reassign or page. Each is a separate tool, granted by name per team, used only when somebody asks for exactly that.
Does it replace our documentation?+
No, it reads it. Confluence, Notion, Guru, SharePoint and Drive are searched with everything else, and what gets worked out in a ticket or a channel is kept as memory, so the second time is faster than the first.
Can we show a client what it used?+
Every answer names the tickets, alerts and records behind it, every scheduled job keeps the report of each run in a chat, and usage is exportable by person, team, model and day.
The same work, by team
The same connectors and the same memory, described through the job somebody does rather than the business they do it in.
- SupportAnswer tickets with the whole company behind youSupport teams spend their day switching tabs: the ticket in Help Scout, the customer in the CRM, the order in the warehouse, the invoice in Xero. Memtro reads the ticket and does the switching for them.Memtro for support
- EngineeringKeep decisions from evaporatingDecisions, owners and deadlines are extracted from conversations and tool results into a typed, time-aware memory. Corrections supersede old facts instead of piling up next to them.Memtro for engineering
- OperationsMonitors and digests that run themselvesJobs are prompts the agent runs on a schedule with the same tools as their owner. Create them in plain language, get the report in a chat, pause or change them in a sentence.Memtro for operations
Ask it about your own business
Connect the first two systems, ask the question you opened this page with, and see where the answer comes from. About five minutes.
7-day free trial · card on file, nothing charged until it ends · cancel any time