01 · Automations & AI agents

Automations that keep working after the demo.

Getting an AI to do something once is easy. Getting it to do the right thing every day, at a predictable cost, and to tell you when it isn't sure, is the actual job. I design and build automations and AI agents for teams that need them to be dependable, not impressive for a week.

Sound familiar?

Signs this is for you

  • 01

    Someone on your team spends hours a week moving information between tools by hand.

  • 02

    Requests and issues arrive in chat and get lost, or nobody knows who owns them.

  • 03

    You've tried an AI tool, and it worked in the demo but not on real data.

  • 04

    You're growing, and want the right systems in place before the busywork piles up.

What you get

How I approach it

  1. 01

    Map it before automating it

    We walk through the process as it really works, not as the diagram says. Most savings come from fixing the process; automation locks it in.

  2. 02

    Rules in code, judgement in the model

    Anything predictable is plain code: free, fast and testable. The AI is only used where judgement is needed, and only woken up when there's something new to judge. That's how one agent's running cost dropped from £490 to £39 a month.

  3. 03

    Built to fail safely

    Outages, missed messages and API limits are designed for from day one. If something goes wrong, the system catches up on its own and never quietly drops work.

  4. 04

    Yours to run

    Plain-English specs, a runbook, and a handover session. Every rule is written down with the reason it exists, so your team isn't dependent on me.

In practice

What this looks like, from simple to serious.

Workflow
Connects the tools you already use, so information moves without being copied by hand.
Automation
Carries out repetitive steps on a schedule, or the moment something happens.
AI agent
Applies judgement where rules can't: reading, classifying, drafting. Always with clear limits and a person in control.
01

Quick wins

Small, fast to set up, built on the tools you already use.

  • Automation

    Bookings that run themselves

    Clients book online; confirmations, reminders and calendar entries follow automatically. No back-and-forth to arrange a time.

  • Workflow

    Every enquiry captured and answered

    Enquiries from your website and social channels receive a prompt, well-written acknowledgement and are recorded in one place, with a follow-up if there's no reply.

  • Automation

    Reviews requested at the right moment

    After each completed job, satisfied clients are invited to leave a review. Steady reputation-building, with no effort from you.

02

Agents in production

Not demos: masked, cost-controlled, tested and observable from day one.

  • AI agentDelivered

    Issue-desk triage

    A report raised in chat becomes a complete, routed ticket within 15 minutes, and every open ticket is re-assessed hourly. Running cost £39 a month, down from £490.

  • AI agentDelivered

    Sensitive data masked before any AI call

    Routine steps are handled by plain rules before a model is consulted, and sensitive data is masked at a single chokepoint before anything reaches an AI model.

  • AutomationDelivered

    Built to recover on its own

    Designed for outages from day one: the system catches up by itself, proven across a 15-hour gap, and never quietly drops work.

03

Platform and scale

Turning one successful agent into a way of building agents.

  • AI agentDelivered

    Research that cites its sources

    Market, customer and compliance research run in parallel. Every claim is cited and dated, disagreement is surfaced rather than smoothed over, and severe findings interrupt a person.

  • WorkflowDelivered

    A shared, governed tool plane

    A catalogue of permissioned tools across Asana, Jira and Slack that any approved agent can call. Built once, shared by every agent on the platform.

  • Automation

    Quality as a release gate

    Automated tests gate every release, and go/no-go decisions are recorded with the evidence behind them.

Proof

Where I’ve done this

Questions

Before you ask

Which tools do you work with?

Slack, Jira, Asana, CRMs, spreadsheets, data warehouses, and most things with an API. If it can't be reached properly, I'll say so before we start rather than after.

Is it safe to put our data through an AI model?

It depends on the data and the model provider, and it's decided up front. Customer data stays out of prompts by default, and anything that must go in is masked or covered by the right agreements.

What does it cost to run afterwards?

You get a running-cost estimate before the build and a measured figure after. The design aim is that cost grows much more slowly than volume.

Contact

Tell me what you need.

A few lines is enough. Every enquiry comes straight to me, and you’ll get a considered reply: a straight answer, a question or two, or an honest “I’m not the right fit”.

Prefer email? hello@firstniyyah.com