Career resource

10 AI Tasks You Can Practise Using Your Current Job

Turn familiar work into safe AI practice by improving a process, documenting a decision, or testing an output without exposing confidential data.

Last updated
2026-09-14

Use familiar work as your starting point

You do not need to invent an impressive fictional problem to practise applied AI. Your current work already contains requests, handoffs, recurring documents, decisions, quality checks, and delays.

Choose a low-risk example and remove confidential or personal information. The goal is to practise the method, not to send restricted work to a public AI tool.

Ten practical tasks

Try one of these bounded exercises:

Map a repetitive request from intake to completion.
Compare a manual process with a rules-based approach.
Draft a structured template for a recurring handoff.
Create a quality checklist for an AI-assisted output.
Identify missing fields that slow down a team decision.
Design a routing rule for incoming work.
Write test cases for an automation or classification step.
Create a simple monitoring plan for an important workflow.
Record the risks and owner for a proposed AI use case.
Rewrite a process explanation so another person can follow it.

Turn practice into honest evidence

Keep the problem, scope, decisions, example artefact, and limitations together. Do not claim that a practice exercise improved performance unless you measured it in the workplace.

Iteretta helps learners develop this habit through practical tasks, structured Hiri feedback, and business challenges.

Common questions

Can I use real work for an AI portfolio piece?

Only when you have permission and can protect confidential, personal, and commercially sensitive information. Otherwise, recreate the process with fictional or fully anonymised inputs.

Which task should I start with?

Choose a frequent, bounded, low-risk workflow where the input, decision, and desired output are clear. Avoid starting with a high-consequence decision.

This resource is maintained by Iteretta. It is educational information, not legal, financial, medical, employment, or other professional advice.

Continue learning

Applied AI Interview Questions for Career Changers ↗How to Make Your CV Less Generic in Five Minutes ↗Tell Me About Yourself: A Practical Structure for Career Changers ↗Start with the free lab preview ↗Explore the relevant lab ↗See lab pricing and access ↗

Use familiar work as your starting point

You do not need to invent an impressive fictional problem to practise applied AI. Your current work already contains requests, handoffs, recurring documents, decisions, quality checks, and delays.

Choose a low-risk example and remove confidential or personal information. The goal is to practise the method, not to send restricted work to a public AI tool.

Ten practical tasks

Try one of these bounded exercises:

Map a repetitive request from intake to completion.

Compare a manual process with a rules-based approach.

Draft a structured template for a recurring handoff.

Create a quality checklist for an AI-assisted output.

Identify missing fields that slow down a team decision.

Design a routing rule for incoming work.

Write test cases for an automation or classification step.

Create a simple monitoring plan for an important workflow.

Record the risks and owner for a proposed AI use case.

Rewrite a process explanation so another person can follow it.

Common questions

Can I use real work for an AI portfolio piece?

Only when you have permission and can protect confidential, personal, and commercially sensitive information. Otherwise, recreate the process with fictional or fully anonymised inputs.

Which task should I start with?

Choose a frequent, bounded, low-risk workflow where the input, decision, and desired output are clear. Avoid starting with a high-consequence decision.