Meeting AI Agent

Meeting AI Agent finds the time and the room, prepares a brief from your own reports and turns the written notes into tasks with owners and dates. Your teams spend less time arranging and chasing meetings, and the people who chair them stay in control of what is shared and decided.

Ask for a proposal See the results

Meeting AI Agent today view: the day’s meetings, briefs waiting for approval, action items due and hours in meetings

What it means for your business

−70%

time spent scheduling

Before
Finding a date took days of emails, and clashes appeared after invitations went out.
With Meeting Agent
The agent checks calendars and rooms together and suggests the first slot that works.
100%

of meetings start with a brief

Before
Figures were copied together the night before and read for the first time in the meeting.
With Meeting Agent
Every attendee receives one page with the numbers, papers and open items before the meeting.
92%

of actions closed on time

Before
Actions were written in the minutes and followed up from memory, if at all.
With Meeting Agent
Each action becomes a task with an owner and a date, and the agent reminds the owner before it is due.

What your managers and teams get back

01

Dates found in minutes, not days

The agent checks everyone’s calendar and your room booking system and proposes a time that works, without the back-and-forth emails.

02

Meetings that start from the facts

A brief with the latest figures, documents and open tasks reaches attendees before the meeting, so time goes on decisions instead of reading numbers aloud.

03

Every decision turns into action

Written notes shared after the meeting become tasks with owners and due dates in Jira or Asana, the same day they are agreed.

04

Nothing quietly falls through

Owners are reminded before deadlines, and anything still open goes onto the next meeting’s agenda automatically.

What it looks like in practice

An office manager asks for a planning meeting with six regional leads. The agent finds a slot that suits everyone, books a room and prepares the brief. The invitations wait in Outlook, ready to send.

Office managerSet up the Q1 planning meeting with the regional leads.

Meeting AgentAll six leads are free for 90 minutes on 14 January. I’ve booked the Atrium room and attached a brief with Q4 results and open projects. The invitations are in Outlook.

Looked up

Room booking
Project system
Outlook
Notion

How six organisations use the Meeting AI Agent

Each one in short: the problem, the results, and how the agent handles the work.

Every board pack out 7 days before the meeting

Two people in the board office spent most of the fortnight before each board meeting chasing papers by email.

Retail banking · Enterprise, on their own servers · Live in 7 weeks

Diagram: where Meeting Agent sits. Inside the bank, the board office and the company secretary ask the agent. It reads the finance reports, the risk register and the decisions log, tells directors in Microsoft Teams, follows up resolution owners and hands the approved pack to Document Agent. The agent provider is outside the bank and only sends updates in.

1
Plans the board cycle backwards
From the board date, the agent books paper deadlines in each contributor’s calendar and reminds them in Teams, so papers arrive on time without a chase.
2
Builds the pack from the bank’s own figures
Capital, liquidity and credit figures come straight from the finance reports and the risk register, each with its source and date.
3
Sends it seven days ahead, after the secretary’s OK
The company secretary checks the draft.
4
Turns approved minutes into tracked resolutions
From the minutes the secretary writes, the agent logs each resolution in the decisions log with an owner and a date, and follows up until it is closed.

7 daysevery pack out on time
−60%time spent assembling a pack
100%resolutions with owner and date
0board papers leaving the bank

Weekly site coordination cut from 90 minutes to 55

Site managers wrote notes during the walk and typed them up when they could, often days later.

Construction · Cloud, run for them · Live in 4 weeks

Diagram: where Meeting Agent sits. The agent runs in a private cloud for the contractor. Site managers and planners ask the agent. It reads the look-ahead plan and written site notes, looks up Asana and Dropbox, sends actions to subcontractors in Microsoft Teams, questions to the design team and follow-ups to Task Agent.

1
Prepares Monday’s agenda
The agent builds the agenda from the three-week look-ahead, last week’s open items and any drawing revisions since the last meeting.
2
Turns written site notes into actions
After the walk, the site manager types short notes on a tablet.
3
Sends every subcontractor their actions
The same day, each subcontractor lead gets their actions in Microsoft Teams, and the actions appear in Asana with the meeting linked.
4
Checks on Friday and carries over
On Friday the agent checks what is done, reminds owners once, and puts anything still open at the top of next Monday’s agenda.

−45%actions carried over unfinished
55 mincoordination meeting, down from 90
Same dayactions reach every subcontractor
2 hsaved per site manager each week

Committee papers out 10 days before every meeting

Committee secretaries spent up to three days finding a date for each meeting.

Higher education · Cloud, run for them · Live in 5 weeks

Diagram: where Meeting Agent sits. The agent runs in a private cloud for the faculty. The faculty office and committee chairs ask the agent. It reads the teaching timetable and room booking, looks up Google Calendar and Notion, sends agendas and papers to committee members in Slack and to the faculty board, and passes decisions to Knowledge Guide.

1
Reads calendars and teaching together
The agent combines personal calendars with the teaching timetable, so a lecture blocks the time even when it is not in someone’s calendar.
2
Proposes times that respect the rules
It suggests two or three times when everyone is free, a suitable room is available and nobody loses a research day.
3
Collects agenda items and builds the agenda
Members submit items in Notion or Slack.
4
Sends papers ten days ahead and logs decisions
Approved papers go out on time.

−90%time to find a committee date
0clashes with teaching
10 dayspapers out before every meeting
4 hsaved per committee secretary a month

20 minutes of every department heads meeting given back

An operations analyst pulled figures from four systems late every Monday evening.

Hospital · Enterprise, on their own servers · Live in 8 weeks

Diagram: where Meeting Agent sits. Inside the hospital, the operations manager and the medical director ask the agent. It reads totals from bed management, the staff rota and waiting lists, shares the brief with department heads in Microsoft Teams, tracks actions in SharePoint and hands trends to Data Analyst. The agent provider is outside the hospital and only sends updates in.

1
Collects the overnight totals
At 05:30 on Tuesday the agent reads bed occupancy, open shifts, waiting list changes and new incidents for each department, as totals only.
2
Writes one page per department
Each department head gets the same layout: four numbers with last week’s values, anything over its threshold flagged, and open actions from last time.
3
Shares it at 06:30, after a person’s OK
The operations manager checks the brief at 06:15.
4
Follows up what was agreed
From the written minutes, actions go to their owners in Teams and are tracked in SharePoint.

−70%time preparing the numbers
1page per department, every week
20 minof each meeting given back
0patient records in any brief

Partners spend 6 fewer hours a week in meetings

Each partner spent about 26 hours a week in meetings, a third of them internal status updates.

Management consulting · Cloud, run for them · Live in 3 weeks

Diagram: where Meeting Agent sits. The agent runs in a private cloud for the firm. Partners and engagement leads ask the agent. It reads team calendars and engagement files, looks up Notion and Calendly, sends internal actions to consultants in Jira, approved follow-up emails to clients in Outlook and client follow-ups to Sales Agent.

1
Finds times with clients and the team
The agent offers clients times through Calendly and finds slots in Outlook for the team, keeping two focus blocks a day free for every partner.
2
Prepares every steering brief
Two days before each steering meeting it builds the brief from Jira status, the engagement page in Notion and open actions, for the engagement lead to check.
3
Turns notes into actions for both sides
From the written notes, internal actions go to Jira straight away.
4
Suggests which meetings to cut
Every Friday it lists recurring meetings without decisions or actions lately, and suggests replacing them with a written update.

−6 hin meetings per partner each week
11of 11 steering meetings with a brief
92%of actions closed on time
Same dayfollow-up email to clients

100% of committee agendas published on time

Reports often arrived on deadline day without a financial or legal section, and the team spent the publication week chasing authors by email and chat.

Local government · Enterprise, on their own servers · Live in 9 weeks

Diagram: where Meeting Agent sits. Inside the council, democratic services and report authors ask the agent. It reads case management, the document archive and the decision register, sends reports to legal for sign-off, actions to departments and financial sections to Finance Agent. The agent provider is outside the council and only sends updates in.

1
Tracks every report against its deadline
The agent lists the reports due for each committee and reminds authors in Teams two weeks, one week and two days before the deadline.
2
Checks each report is complete
It checks for a summary, financial implications and legal sign-off, and sends incomplete reports back to their author with what is missing.
3
Builds the agenda and papers
Complete reports become the draft agenda and papers pack, ready for democratic services to check and publish seven days ahead.
4
Records decisions and sends actions
Once the minutes are approved, the agent records each decision in the decision register and sends actions to the departments responsible.

100%agendas published on time
−50%late or incomplete reports
1 dayfrom approved minutes to actions
0records leaving the council

Illustrative scenarios with unnamed companies. The figures show the kind of improvement each setup is designed to deliver; actual results depend on volume, the systems connected and how the agent is configured.

From first conversation to live in three to nine weeks

We set up the Meeting AI Agent around how your managers, coordinators and board or committee secretaries run meetings today.

1

Week 1

Learn your process

We sit with the people who do the work, map the process step by step and find the systems that hold the information. You get a written plan before anything is built.
2

Weeks 2 to 3

Customize your agent

We build and configure the agent around your process: your rules, your approvals, your wording, and connections to your databases and the tools your teams use.
3

Weeks 3 to 4

Deploy and test with your team

We install it inside your infrastructure, or in our cloud if you prefer, and test it with your team on real requests until the results meet the standard you set.
4

From week 5

Onboard your team

We train your employees and administrators, hand over the controls and stay with you: monthly reviews, updates you approve, and the next process when you are ready.

What your team and managers see

Meeting AI Agent, today screen — **Today.** Your meetings, what still needs preparing and what is waiting for your OK.

Meeting AI Agent, meeting briefs screen — **Meeting briefs.** One page per meeting with the agenda, the figures that matter and open items, prepared from your own systems.

Meeting AI Agent, action items screen — **Action items.** Everything agreed in your meetings, with owners, due dates and the reminders the agent has sent.

Meeting AI Agent, the agent’s own role screen — **The agent’s own role.** What the agent may do alone, such as booking rooms or posting tasks, and what always needs a person’s OK.

Meeting AI Agent: questions from business leaders

How quickly will we see results?

Your team tests the agent on real meetings within the first weeks. Most organisations see the time saved on scheduling and preparation, and more actions closed on time, during the first month after going live.

What does it cost?

It depends on how many people and meetings it supports, the systems it connects to and where it runs. After a first conversation you get a written proposal with the price, so you can compare it with what the work costs you today.

Will it replace our coordinators and secretaries?

It takes over the chasing, copying and reminding, so the same people can support more meetings and spend their time on the content. What goes on the agenda, what is shared and what is decided stays with your people.

What if a brief contains a wrong figure?

It takes figures only from your own reports and systems, shows the source and date of each one, and nothing is shared until the person who owns the meeting approves it. Every brief and action is logged, so you can review and correct it.

Can it reschedule meetings on its own?

Only if you allow it. Otherwise it suggests new times and the organiser confirms. You decide what it may book, send and post alone, and you can pause it at any time.

Is our information safe?

It can run on your own servers, with its AI models, so board papers, notes and figures stay inside your organisation and no outside AI service sees them. We have no access to it, and your data is never used to train AI models.

Do we have to change our calendars or tools?

No. It works with Outlook and Google Calendar, your room booking system, Microsoft Teams, Slack, Jira, Asana, Notion and Confluence.

Does it sit in on our meetings?

No. It never joins a meeting. It works in writing only: calendars, briefs, documents, the notes your team shares afterwards and the tasks that follow.

What do you need from our team?

One person who knows how your meetings are organised today, access to test versions of the calendars and systems involved, and a few hours a week from the team while we test.

Can we start small?

Yes. Most organisations start with one recurring meeting, such as a management or committee meeting, see the results, and then extend the agent to other teams.

See what the Meeting AI Agent can do for your management team.

Tell us which meetings take the most preparation, where the figures and papers come from and what happens to actions afterwards. You get a written proposal with the expected results, the setup, a timeline and a price.

Ask for a proposal info@ihtech.ba

Sarajevo, Bosnia and Herzegovina