Use case library
AI use cases,
worked through.
Each page is one specific business problem and the system I would build to solve it: what sets it off, what the model actually does, what stays plain code, which of your systems it connects to, and where a person stays in the loop. AI agents, automated workflows, integrations and custom software, for the software you already run.
They are worked examples, not client stories. Where a page draws on something I have actually shipped, it says so. Everything else is exactly how I would build it for you.
- Use cases
- 60
- Functions
- 10
- Industries
- 14
- Systems
- 15
By what your team does
Start from the function where the time goes.
Customer support
Tickets, chats and inboxes answered from real data, with a person on anything that costs money.
5 use cases →
Sales and lead handling
Leads answered, researched and routed in minutes, and quotes drafted before anyone retypes them.
7 use cases →
Finance and accounting
Invoices, e-invoicing, collections and reconciliation, with a person on anything that moves money.
6 use cases →
Operations and logistics
Orders, inboxes, suppliers, dispatch and paperwork: the coordination work between your systems.
9 use cases →
Document processing
PDFs, scans and attachments turned into checked, structured data your systems can use.
9 use cases →
HR and recruiting
Screening, onboarding and redeployment, designed so a person makes every decision about people.
3 use cases →
Internal tools and knowledge
Company knowledge, reports and safe AI access for staff, built on your data and your permissions.
6 use cases →
Marketing and content
Catalogs, imagery, returns and reviews at scale, with a person approving claims and brand.
5 use cases →
AI in your product
In-app assistants, MCP servers, portals and catalogs: AI your customers use, not only your team.
4 use cases →
Modernization and rescue
Broken automations, spreadsheet operations, legacy code and prototypes, made dependable.
6 use cases →
By industry
Workflows that only exist in your line of work.
By the software you run
Nothing gets replaced. The build sits across these.
Start with these
The problems I hear most often, and the builds that pay for themselves fastest.
Customer orders from email, PDF and WhatsApp, entered into the ERP without retyping
Turns emailed PDFs, Excel forms and WhatsApp photos into sales orders in SAP Business One or NetSuite, and sends only the doubtful lines to a rep.
Outlook and Microsoft 365 / WhatsApp Business Platform / SAP Business One / NetSuite / Excel order forms
Supplier invoices from inbox to ledger, with the checks a good controller would do
Reads supplier invoices from the AP inbox, matches them to purchase orders and goods receipts, and holds any invoice whose bank details changed until someone calls.
Outlook (Microsoft 365) / Xero / QuickBooks Online / DATEV / NetSuite
A support agent that answers 'where is my order?' from live Shopify and carrier data
Answers order questions on every channel from live Shopify and carrier data, verifies the customer first, and hands refunds and angry customers to a person.
Shopify / Gorgias or Zendesk / WhatsApp Business Platform / Carrier tracking API / Returns app (Loop, ReturnGO)
A company assistant that answers from SharePoint and Drive, and respects who may see what
Answers staff questions in Teams and Slack from SharePoint, Drive and Confluence, cites every source, and only searches what the person asking may open.
SharePoint and Teams / Google Drive / Slack / Confluence or Notion / Microsoft Entra ID or Google groups
Every inbound lead researched, scored and routed before a rep opens it
Researches and scores every inbound lead against your rubric, catches existing customers, routes by territory and capacity, and replies with a booking link.
HubSpot / Salesforce / Enrichment provider (Clay or Apollo) / Cal.com or Calendly / Slack
An agent for the shared inbox: every email sorted, answered or filed into the right system
Reads every email in your shared mailboxes, sends orders and invoices into the right system, drafts replies for approval, and flags fake bank-detail changes.
Microsoft 365 shared mailboxes / Gmail and Google Workspace / The ERP (Business Central in this example) / Teams or Slack / The AP flow and accounting system
Connect your ERP and CRM to Claude, ChatGPT or Copilot without handing over the keys
Gives the assistants your staff already use a few scoped tools into the ERP, CRM and databases: read-only by default, writes behind approval, every call logged.
SAP Business One / HubSpot / Postgres or SQL Server database / Claude, ChatGPT and Copilot Studio / Microsoft Entra ID or another OAuth provider
Fixing n8n, Zapier and Make automations that report success and did nothing
Audits inherited n8n, Zapier and Make automations, then adds a run ledger, idempotency keys and a daily reconciliation so a green run means the work was done.
n8n / Zapier / Make / Postgres / Slack
Every use case
A second-line support agent that checks the logs before an engineer gets pulled in
Investigates escalated tickets the way an engineer would, from account settings to error logs to known issues, and returns a checked answer or a ready bug report.
A support agent that answers 'where is my order?' from live Shopify and carrier data
Answers order questions on every channel from live Shopify and carrier data, verifies the customer first, and hands refunds and angry customers to a person.
One support team answering customers in twelve languages without hiring for each one
Lets a small team work in its own language while customers get replies in theirs, with product names protected, formality per market and legal wording pre-approved.
Every support conversation scored against your own rubric, not a two percent sample
Scores every conversation against your rubric with quoted evidence, calibrated against your QA lead, and built for coaching rather than discipline.
Warranty claims triaged from photos, serial numbers and receipts before anyone opens them
Reads the serial off a label photo and the date off the receipt, checks both against your ERP and warranty terms, and hands a person a claim ready to decide.
Every inbound lead researched, scored and routed before a rep opens it
Researches and scores every inbound lead against your rubric, catches existing customers, routes by territory and capacity, and replies with a booking link.
Freight quote requests in four languages, answered in minutes and priced by your own rules
Reads freight quote requests in your customers' languages, asks for what is missing, prices the lane with your margin rules, and sends or drafts the quote.
Law firm intake that runs the conflict check before anyone calls the prospect back
Captures every inquiry, after hours too, searches conflicts across parties and their groups, flags deadlines, and prepares the engagement letter for a lawyer.
RFQ emails with drawings turned into draft quotes in your ERP, priced from your own job history
Reads RFQ packages, measures STEP files, flags revision and unit conflicts, finds similar past jobs and drafts an ERP quote for the estimator to price.
An after-hours phone agent for trades that books jobs and puts emergencies through
Answers after-hours calls, puts gas and water emergencies through to the on-call technician, books routine jobs into real slots, and transfers anyone who asks.
Finding public tenders worth bidding on, and answering them from your own bid library
Watches TED and national portals, scores tenders against your capabilities, builds the compliance matrix, and drafts answers only from approved sources.
Sales calls and emails that update the CRM on their own, after the rep approves
Reads customer calls and email threads, proposes stage, next step and close date changes with evidence, and writes them to the CRM after one click from the rep.
An e-invoice layer that lets an old ERP issue XRechnung and ZUGFeRD before 2027
Reads invoice data from the ERP you already have, maps it to EN 16931, generates XRechnung or ZUGFeRD, validates every file with KoSIT and sends it by email or Peppol.
Receiving XRechnung, ZUGFeRD and Peppol invoices without retyping a single field
Structured e-invoices end the retyping, not the work: this pipeline validates them, renders them for approvers, matches them to orders and archives the original XML.
Supplier invoices from inbox to ledger, with the checks a good controller would do
Reads supplier invoices from the AP inbox, matches them to purchase orders and goods receipts, and holds any invoice whose bank details changed until someone calls.
3PL invoices built from warehouse events instead of month-end spreadsheets
Computes storage, pick and pack and value-added service charges from WMS events and each client's rate card, drafts invoices, and explains any line a client questions.
Chasing clients for receipts and bank statements without an accountant doing the chasing
A per-client checklist of what should arrive each month, polite reminders that name exactly what is missing, and every photo and PDF filed against the right booking.
Collections follow-up that stays polite, logs promises to pay and routes disputes to a person
Staged payment reminders that check for payments first, read every reply, log promises to pay and pause for disputes, with anything final left to a person.
An agent for the shared inbox: every email sorted, answered or filed into the right system
Reads every email in your shared mailboxes, sends orders and invoices into the right system, drafts replies for approval, and flags fake bank-detail changes.
An agent that chases suppliers for delivery dates and updates the ERP when they change
Emails suppliers about open POs in their language, reads confirmations and date changes, writes agreed dates to the ERP and flags the late parts that will hurt.
Customer orders from email, PDF and WhatsApp, entered into the ERP without retyping
Turns emailed PDFs, Excel forms and WhatsApp photos into sales orders in SAP Business One or NetSuite, and sends only the doubtful lines to a rep.
Change orders and delays caught in site logs and emails before the notice deadline passes
Reads site logs, emails and drawing revisions for unpriced changes and hindrances, starts the contract's clock, and hands the PM a notice draft with the evidence.
Grid connection and subsidy paperwork for solar and heat pump installers, prepared by an agent
Fills each grid operator's form from the job file, chases what the operator still wants, and prepares the MaStR entry, with a person checking before anything is sent.
Guest registration, police reporting and tourist tax handled before check-in, in Spain and Italy
Guests share their ID data through a pre-arrival link, the build validates it and reports it to the police systems on time, and the desk only sees the exceptions.
Maintenance requests triaged, dispatched and chased, with a real emergency path
Takes tenant repair requests on any channel, wakes the on-call person only for real emergencies, and dispatches and chases contractors for everything else.
Supplier invoices turned into live recipe costs for a multi-site restaurant group
Every supplier invoice becomes current ingredient prices, so recipe costs stay live, price creep gets caught, and substitutions reach the chef before service.
Cancelled clinic appointments refilled from the waitlist before the slot goes empty
When a patient cancels, the agent offers the slot to waitlisted patients who fit, books the first confirmed reply and tells everyone else the slot is gone.
Catching edited bank statements, payslips and invoices before they cost you money
Cross-checks uploaded statements, payslips and invoices against bank data, VAT records and the file's own structure, then puts the evidence in front of a reviewer.
Commercial insurance submissions assembled from proposal forms, loss runs and property schedules
Reads a client's proposal forms, loss runs and property schedules, reconciles them, and drafts one submission per insurer for the account handler to check and send.
PDFs into rows: document data entry that sends only the doubtful fields to a person
Reads delivery notes, proofs of delivery and forms from an inbox or folder, checks each field against your TMS or ERP, and asks a person only about doubtful values.
Commercial leases turned into key dates and clauses, with a lawyer checking the ones that matter
Turns a portfolio of commercial leases and their amendments into cited abstracts and computed deadlines in Yardi or MRI, with a lawyer confirming breaks and options.
Contract review against your own playbook, with every flag tied to the clause that caused it
Checks each clause of a customer's DPA or NDA against your playbook, quotes the text behind every flag, and drafts tracked changes that a lawyer accepts or rejects.
Court mail read, deadlines calculated by rules, and every one confirmed by a lawyer
Reads incoming court documents, calculates each deadline with tested rules rather than a model, proposes entries in your case system and waits for a lawyer to confirm.
Customs declarations drafted from commercial invoices and packing lists, for the declarant to sign
Cross-checks shipment documents, reuses tariff codes from your accepted history, argues new ones with EU rulings, and drafts declarations for a declarant to sign.
Mortgage applications where the documents chase themselves and the underwriter gets a checked file
Keeps a live checklist per borrower, chases what is missing on their preferred channel, reads and cross-checks each document, and gives the adviser a clean file.
Referral faxes and letters turned into patient records, with missing details chased automatically
Reads each incoming referral, checks it against what the exam needs, links the right patient and chases the referrer for gaps; clinicians keep every clinical call.
Candidate screening that quotes the CV behind every score, built for the EU AI Act
Scores each application against a rubric the recruiter wrote, quotes the CV line behind every score, catches hidden instructions, and never rejects anyone on its own.
Onboarding that provisions accounts only after approval, and answers new starters' questions
Turns a signed contract into an approved access plan, creates the accounts through APIs, answers new starters from the handbook, and reverses it all on the last day.
Redeploying contractors and near-miss candidates before paying to source new ones
Finds contractors rolling off and past near-misses who fit a new job order, drafts personal outreach in the recruiter's mailbox, and updates Bullhorn with every reply.
A company assistant that answers from SharePoint and Drive, and respects who may see what
Answers staff questions in Teams and Slack from SharePoint, Drive and Confluence, cites every source, and only searches what the person asking may open.
A Monday KPI brief that explains last week's numbers instead of just charting them
Computes last week's KPIs in SQL, checks the data is complete, finds what moved and why, and posts a short written brief to Slack at 07:00 every Monday.
Connect your ERP and CRM to Claude, ChatGPT or Copilot without handing over the keys
Gives the assistants your staff already use a few scoped tools into the ERP, CRM and databases: read-only by default, writes behind approval, every call logged.
A private ChatGPT for every employee, behind single sign-on, with redaction and team budgets
A company chat app behind single sign-on that strips client data and secrets, routes to approved EU or self-hosted models, logs use, and gives each team a budget.
Agency client reports with commentary drafted from the data and approved by the account lead
Loads every client's ad and analytics data, measures it against their own targets, names only causes it can prove, and drafts commentary in the client's language.
Plain-English questions over your own database, with the SQL shown and checked
Answers plain-English questions from a read-only copy of your database, shows the SQL and definitions behind each answer, and asks when a question is ambiguous.
Listing one catalog on Amazon, Otto, Kaufland and bol.com without the rejection loop
Turns one PIM catalog into localized, schema-valid listings for Amazon, Otto, Kaufland and bol.com, blocks anything missing GPSR or EPR data, and handles rejections.
On-model photos for every colorway, generated from your packshots and approved by a person
Turns packshots and licensed model references into on-model images across the catalog, measures each against the real garment, and publishes only what a person approves.
What your returns are telling you: return reasons turned into product and size-guide fixes
Reads return comments, reviews and tickets in every language, ties them to SKU, variant and batch, and sends product a weekly list of issues with counts and quotes.
A room visualizer that hangs your real fabrics in the customer's own window
Shoppers upload a room photo, try your fabrics at real pattern scale in their own window, then buy or request a made-to-measure quote that reaches your team in HubSpot.
Replies to every review at every location, and the recurring problems they point to
Collects every location's reviews, drafts replies a manager approves, sends hygiene and safety complaints to a person the same day, and builds a weekly issue list.
A remote MCP server so customers can use your SaaS from Claude, ChatGPT and Copilot
Your product inside Claude, ChatGPT and Copilot Studio: OAuth sign-in, a small set of agent-shaped tools, per-tenant audit logs and rate limits that protect your API.
An in-app AI assistant that acts on the user's own data, not one that only quotes the docs
Tool calling over your own API with the signed-in user's permissions, a confirm step before every write, undo, per-plan usage caps and a trace of every step.
A customer portal where B2B clients track orders, download documents and ask questions
Each customer company gets its own logins, live orders and shipments from the ERP and carriers, every document in one place, and an assistant that sees only their data.
Product data AI shopping assistants can read, filled from supplier specs instead of guesses
Checkout inside chat stalled; discovery did not. Attributes filled only from supplier documents, clean feeds and markup, and a monitor for what assistants say.
An AI bill growing faster than your users: a cost audit, model routing and budgets
Traces model spend to each feature and customer, fixes caching, batching and prompt bloat, routes simple calls to small models, and proves quality holds with evals.
Automating portals and legacy systems that have no API, with a person approving the final click
Keys data into insurer portals and API-less desktop systems with a computer-use model, logs every screen, and waits for a person to approve each submission.
Fixing n8n, Zapier and Make automations that report success and did nothing
Audits inherited n8n, Zapier and Make automations, then adds a run ledger, idempotency keys and a daily reconciliation so a green run means the work was done.
Turning a 40-tab dispatch spreadsheet into an internal tool with rules, roles and history
Moves daily planning off a 40-tab spreadsheet into a web tool with validated forms, roles and a full edit history, while people keep Excel for analysis.
Modernizing a 20-year-old Delphi or VB6 application with AI and tests, one module at a time
Documents what an old application really does, pins that behavior with tests built from real data, then replaces it module by module while it keeps running.
Taking a vibe-coded app to production: secure data, reliable billing, maintainable code
Audits a Lovable, Bolt or v0 app's Supabase policies, secrets and Stripe billing, fixes what exposes data or money first, then adds tests and a deploy pipeline.