Enablement and workshops

AI Agents for Finance Teams

Half a day in which the finance team builds three agents on its own monthly processes.

For FP&A, controlling, accounting and finance operations. No technical background and no code.

What you get

01

Three agents from your processes

Chosen in the scoping session and built in the workshop on the team's own files and examples. Recurring directions: close preparation, replies to vendors and internal requests, variance analysis and explanations, and summaries for leadership.

02

A map of what belongs where

Which finance work belongs to AI, which to automation, and which stays in the core systems. The map that stops you building an agent where a script is needed.

03

A template to continue

How to define an agent: data, rules, examples, and exactly where a person checks the output before it enters the process.

How it works

  1. Scoping session

    We map the systems, data access and monthly processes, and choose three use cases from the team's own work.

  2. Half-day workshop

    Agent fundamentals and what fits finance work, then building the chosen agents on your tools and data.

  3. Weekly support, optional

    Weekly advisory hours until the agents run in the monthly routine and the team continues on its own.

Format

Duration
Half a day, hands-on, after a scoping session.
Participants
Up to 15, in Hebrew or English. Facilitators who speak both finance and AI.
Tools and data
Your approved AI environment, files and examples from the team, within your existing permissions policy. No connection to core systems at this stage.
What we need from you
Two or three practitioners for the scoping session, and examples screened for sensitivity.

FAQ

Yes. All building is done in the approved tools without writing code. What it needs is deep familiarity with the processes, which the team has and no outside consultant does.

Not in the workshop. The workshop works on files and examples. Connecting to a core system is a separate decision with IT, and it is usually better to export the data than to connect to the system.

Automation fits a fixed process with clear rules; an agent fits a task that needs judgment on text, documents and exceptions. Part of the workshop is the map that divides the two.

Contact

Tell us how many people are on the team, which systems you run, and which monthly process hurts most.

The entrance to the ApplAI office at Azrieli Towers, Tel Aviv