Champs AI Lab
Practical AI training for teams and professionals
We teach people to use AI on the work they already do, and to build the tools and agents that keep doing it. On-site in Singapore, for SME teams, small businesses and self-employed agents.
What we teach
Working with messy data
Cleaning, structuring and charting spreadsheets, receipts and system exports without writing formulas.
Drafting to a house format
Quotes, minutes, reports and replies that come out in your format rather than a generic one.
Reading long documents
Contracts, tender packs, email threads and meeting transcripts summarised against the questions you actually have.
Building tools and agents
Turning a task you repeat every week into a saved tool, or an agent that runs it for you, so it is built once rather than done by hand every time.
Checking the output
How to verify what was produced, and the failure modes that matter: invented figures, missing context, confident wrong answers.
Using it safely
PDPA, customer data, what should never be pasted in, and how to write a one-page policy the team will actually follow.
What people use it for
Examples drawn from the functions most businesses run in-house. The same tasks land the same way whether they are split across a team or all handled by one person. Where you provide your own templates and files, we use those instead.
Finance
— Cleaning an aged receivables export and identifying what is genuinely overdue
— Turning a year of job logs into monthly revenue, quote conversion and days sales outstanding
— Receipts and expenses tracked and totalled for reimbursement claims and tax season
— A first pass at variance commentary on management accounts
Operations
— Reconciling job logs, schedules and service records held in different formats
— Turning a thread of site updates into a structured daily report
— Scheduling appointments, viewings and site visits out of a long message thread
— Drafting method statements and job sheets from a standard template
Admin and HR
— Meeting notes into a decisions-and-actions register with owners and dates
— Standard letters, SOPs and onboarding packs kept to one house format
— Summarising policy documents and contracts into something people will actually read
— Client and staff records kept consistent across the files they live in
Sales and marketing
— Quotes, proposals and listing copy drafted from your own template and price list
— Enquiries from web forms, portals and WhatsApp deduplicated, segmented and prioritised
— Client meeting notes turned into a follow-up summary and a next-action list
— A weekly newsletter, market update or post-campaign summary drafted from your own notes
The introductory session
What happens in a 45-minute session
On-site, over lunch, 45 to 50 minutes. No laptops, no prep, nothing to install.
Two demos, run live on screen
Cleaning a messy file
A job log with three date formats in one column, the same client spelled four ways and amounts stored as text, turned into a clean dashboard in about four minutes. No formulas are written.
Building a reusable tool
A saved tool built on screen for a task the team does weekly, then run on two different inputs to show it holds its format.
The session also covers what AI cannot do.
It has no knowledge of your business it has not been given, it will produce confident wrong numbers, PDPA obligations remain yours, and judgment about what good looks like stays with the person who has done the job for fifteen years.
How people adopt AI
01
Use it
Individuals prompt ad hoc, when they remember to. The gains are real, but they leave with the person who was doing it.
02
Build it
Someone turns a task they repeat into a saved tool. Built once, run by anyone on the team.
03
Install it
A team's main recurring tasks are built, documented and run by everyone. This is the stage where the time saved becomes measurable.
Adoption tends to move through these three stages, whether it is a team of thirty or one person working alone. Which stage you are at is what determines the right format below.
What we run
The 45-minute session
On-site · 45–50 min · as many people as the room holds
Two live demos and a plain account of the limits. Suitable for any function; no technical background assumed.
Foundations — half day
On-site · half day · hands-on · up to 15
Hands-on. Participants work on their own files and leave with three working prompts for their own job, plus a clear sense of where the limits are.
AI Implementation for SME Teams
On-site · 4 sessions × 2 hours · up to 15
— Foundations and prompt patterns
— Data work: cleanup, analysis and reporting, on your own files
— Building reusable tools the team keeps
— Rollout and governance, including a one-page AI use policy that respects PDPA
Each participant finishes with a tool they built themselves and a before-and-after measurement on a task they own.
Custom — by role or department
Finance · operations · admin · sales
Built around the files and tasks a single department, or a single self-employed professional, handles day to day.
We scope per company — team size, number of sessions, and whether we work on a prepared dataset or a sanitised file of yours.
Arrange a session or ask a question
Tell us roughly how many people and which functions, and we will suggest a format and a time.