AI discovery
A fixed-fee AI discovery, credited toward your retainer. We map how your work runs and rank where AI could help first.
Start here · Fixed fee, credited toward your retainer · 2–3 weeks
Fractional AI team for Australian businesses · Sydney
Your AI team, without hiring one.
We’re a senior business analyst and an AI engineer. We find the work that slows your team down, build AI that handles it, and keep it running, month after month.
Australian AWS region by default; any exception only with your written agreement.
Fixed fee
Discovery, credited toward your retainer
2–3 weeks
Typically, from kickoff to a ranked backlog
Monthly
Then a retainer · Quoted after discovery
Australian AWS region
By default for client builds
Big companies have AI teams. You have a busy team.
Most growing businesses can see what AI could do. The hard part is finding the time to pick the right tools, set them up properly and keep them working.
How discovery finds where the time goesRepetitive work buried in inboxes
Invoices, forms and enquiries are re-keyed and chased by hand across inboxes, documents and spreadsheets.
AI tools, and nobody to set them up
There are plenty of AI tools to try, but nobody has the time to choose one, set it up and check what it produces.
Projects that stall
A one-off project looked good at the demo, then stalled once the consultant moved on.
How we work with you
Start with discovery. Then we work through your backlog, one process at a time.
Fractional AI team
A monthly retainer that works through your AI backlog one process at a time: pick, build, test, train and measure. It runs after a discovery.
Monthly · Quoted after discovery
AI document processing
Reads supplier invoices, purchase orders, forms and contracts, checks them against your rules and flags exceptions for a person to decide.
Part of the monthly retainer
Inbox and enquiry triage
Sorts shared inboxes and web enquiries, pulls out the key details, routes each one to the right person and drafts a reply for them to approve.
Part of the monthly retainer
Report and summary drafting
Drafts weekly and monthly reports from your own systems, with every line linked to its source and anything unsupported flagged for review.
Part of the monthly retainer
Knowledge assistant
Answers staff questions from your own policies, procedures and product information, citing the document and section, or saying when it can’t find one.
Part of the monthly retainer
AI voice agent
Answers the calls your team misses after hours or at busy times, says it’s an AI assistant, handles simple requests and passes the rest to a person.
Part of the monthly retainer
What it looks like in practice
Three illustrative examples of the kind of output we build. The businesses, people and documents are fictional.
Illustrative example· fictional invoice Illustrative example of document checks for a fictional supplier invoice. A unit price on line 3 differs from the purchase order and is rated high. The supplier ABN matches the vendor record and the GST amount is 10 per cent of the subtotal, both rated low.
An invoice check against the purchase order
For a fictional supplier invoice. Each check names the rule it applied, and a person decides on the exceptions.
AI document processingIllustrative example· fictional customer Illustrative example of an enquiry record for a fictional customer, J. Citizen, asking about a delivery for order SO-2291. The type is delivery enquiry, urgency is normal and it is routed to customer service. A drafted reply is waiting for approval and nothing is sent until a person approves it.
A customer enquiry sorted and routed
A fictional delivery enquiry turned into a structured record, with a drafted reply that waits for a person to approve it.
Inbox and enquiry triageIllustrative example· fictional business Illustrative example of a weekly operations summary for a fictional business. Three lines link to a sales export, a dispatch log and a CRM report. One line about a stock count variance has no source and is flagged for review.
A weekly summary, linked to its sources
For a fictional business. Each line links to where it came from; any without a source are flagged for review.
Report and summary drafting
How it works
Three stages. You know the scope and price before each one starts, and you decide whether to go on to the next.
Discovery
Fixed fee · 2–3 weeks
We map how work flows with the people who do it, then give you a ranked backlog of what AI could take on. You decide whether to go further.
Led by our business analyst
Monthly cycle
Each month, on the retainer
We pick the next process, build it, test it against your own examples, train the team who will use it, and measure how it is going.
Led by the whole team
Ongoing
Alongside every month
We monitor quality, keep an eye on running costs, and send a short monthly report on what shipped, what is in progress and what is next.
Led by our AI engineer
Security
Built for confidential business information
Where your data goes and who checks the output are the first questions to ask about AI. These are the commitments we design every build around. The binding terms are in your client agreement.
Australian AWS region by default
For client engagements we build on AWS in Australia (Sydney region where available); any exception only with your written agreement.
Built so a person reviews outputs before they go out
Drafts go to a review step designed with the person who knows the work, who approves, edits or rejects them.
Built to show its sources
Results link to the document, email or record they rely on. Any without a source are flagged for review, not hidden.
Built with a plain-English system description
Each build comes with a short description of what it reads, what it produces, where a person reviews and where it runs.
Systems you already use, such as Microsoft 365 or Google Workspace, keep their data where their providers keep it, under your agreements with them.
Who you’ll work with
A senior business analyst leads discovery and adoption. An AI engineer leads the build. You work with the whole team throughout.
About usSenior business analyst
Discovery and adoption
- Senior business analyst and product owner, 12+ years in digital transformation
- Leads discovery, process mapping, adoption and training
- Master of Technology Management (UP Diliman) · PSPO I · ITIL Foundation
AI engineer
Build, testing and running
- AI engineer, 12 years in software, six building production AI agents (retrieval, guardrails, evaluation)
- Builds, tests and runs the systems
- BSc Information Technology (University of the East)
Questions people usually ask
Something else on your mind? Email hello@habiworks.com.
Do we need a tech team?
No. We need one owner who decides priorities and one champion in each team who tests the work and helps teach it. We set everything up and run it with you, and work with your IT provider if you have one.
What do you need from us?
Time from the people who do the work during discovery (agreed up front), access to examples of the documents or emails involved, and someone who can decide what comes next each month.
Which tools do you use?
Amazon Bedrock on AWS by default, working inside the tools your team already uses. We agree the tools during discovery.
Where does our data go?
For client engagements we build on AWS in Australia (Sydney region where available); any exception only with your written agreement. Systems you already use, such as Microsoft 365 or Google Workspace, keep their data where their providers keep it, under your agreements with them.
Can we pause or stop the retainer?
The terms for pausing or stopping are agreed in your client agreement.
How is the retainer priced?
Quoted after discovery, based on how much we take on each month. Discovery has a fixed fee, agreed before it starts and credited toward your retainer.
Next step
Tell us what’s slowing your team down.
A fixed-fee discovery, credited toward your retainer. A 15-minute call first, with no preparation and no client data.