AI agents for small local businesses
Your business,
on autopilot.
We build AI agents that answer enquiries, book appointments, prepare quotes, and keep paperwork up to date. You get more time for work that cannot be automated.
ncode — agent session
$ ncode run booking-agent
→ reading inbox & DMs… 14 booking requests
→ checking the calendar… done
→ confirming & sending reminders… done
→ flagging 1 conflict for you
✓ 13 bookings confirmed · 2h 10m saved
$
What we build
// capabilities/01Autonomous agentsYou set the goal and the rules. The agent plans each step, works in the tools you already use, and checks the result before sending anything.
/02Workflow automationThe agent handles bookings, reminders, quotes, listings, and follow-ups when they are due. You approve only the decisions that need your input.
/03Works with your existing toolsWe connect the agent to your email, calendar, messaging, and payment tools. You keep working the same way, without another dashboard to check.
Watch an agent work
// live demoThe agent reads the task, completes it in the right tools, and checks the result. The demo lets you follow each step.
src: inbox & DMs
BK-0412
read_dm
calendar
send_sms
{}AGENT
dst: calendar booked
BK-0412✓
BK-0412✓
BK-0412✓
⚑ you approve
conflicts go to you
BK-0412
BK-0412✓
BK-0412
BK-0412✓
BK-0412
BK-0412✓
BK-0412
BK-0412⚑
$ booking-agent: reads requests → checks the calendar → confirms & reminds · conflicts go to you
How it works
// how it works/01
Audit
Together, we find the tasks taking up the most time. After the free audit, you get a ranked list of processes worth tackling first.
/02
Prototype
Within 2–4 weeks, we run a prototype on your real data. You can test it before committing to a full deployment.
/03
Deploy
We connect the agent to your tools, deploy it securely, and monitor it from day one.
/04
Improve
After launch, we teach the agent how to handle unusual cases. Once it runs reliably, we can give it another process.
Where agents earn their keep
// examplesHair salon
Fewer gaps in the booking calendar
The agent answers booking requests from calls, messages, and email, then confirms appointments and sends reminders. The calendar has fewer gaps, and bookings no longer take up the evening.
$ ncode run booking-agent
✓ 14 bookings confirmed · 9 reminders sent
Car detailing
Quotes ready in minutes, not days
The agent prepares a quote from the customer's photos and your price list. It is ready for your approval within minutes, rather than days.
$ ncode run quote-agent
✓ 11 quotes sent · 3 jobs booked
Electronics resale
From buy-in to a live listing
The agent identifies new stock, checks market prices, and publishes a listing with photos and a description. You can stay at the repair bench.
$ ncode run listing-agent
✓ 23 items listed · 4 prices updated
works with the tools you already use · Gmail ✦ Google Calendar ✦ WhatsApp ✦ Instagram ✦ Booksy ✦ Allegro ✦ OLX ✦ Google Sheets ✦ Stripe ✦ QuickBooks ✦ Slack ✦ Notion ✦ works with the tools you already use · Gmail ✦ Google Calendar ✦ WhatsApp ✦ Instagram ✦ Booksy ✦ Allegro ✦ OLX ✦ Google Sheets ✦ Stripe ✦ QuickBooks ✦ Slack ✦ Notion ✦
// do the math
What's the boring work costing you?
Move the sliders to estimate the savings. The calculator assumes an agent handles about 70% of the repetitive work. For a clearly defined process, that is a conservative estimate.
Hours per week on repetitive work40 h
Average hourly cost$45 /h
Hours back per year
1,456 h
Cost recovered per year
$65,520
≈ full-time roles worth of time0.8
You stay in control
// guardrailsHuman approval
You approve what matters
You decide which actions the agent can take and which need your approval. Anything irreversible always waits for your sign-off.
Your cloud
Runs where you say
The agent can run in your environment or ours. It gets only the access it needs in each system.
Audit log
Every action, logged
We record every step, tool call, and decision. You can see what the agent did and why.
No training
Your data stays yours
Your data is never used to train models. Contractually, not just as a promise.
Questions
// faq// see a prototype, not a slide deck
See an agent handle one of your tasks.
Describe a task that takes time every week. We will tell you whether it is worth automating and show you with a working prototype.
hello@ncodelab.com