Skip to content
Christy BuildsWork with me

Find the work AI should be doing for you.

I help organizations find their highest-friction workflows, work out where AI and automation would create measurable value, and build practical systems that fit the way the business actually runs.

01Where it usually starts

Most of the cost hides in ordinary work.

It turns up as hours, delays, and the one person everyone has to ask.

  • Your team copies information from one system into another by hand.
  • Important processes live in spreadsheets, inboxes, and one person’s memory.
  • People spend hours processing documents, updating records, or rebuilding the same report.
  • Your CRM and your other systems don’t talk to each other, so people do it for them.
  • Requests get routed, chased, and followed up manually.
  • You know AI could help. You don’t know which ideas are worth building.
02How I work

Understand the work first. Build second.

Most of the value is decided before anything gets built. Half the time the real fix is cleaning up the process underneath, and a model comes after, if it belongs at all.

  1. 01

    Understand the current workflow

    How the work actually happens, including the workarounds nobody wrote down.

  2. 02

    Find the friction

    Where the time, cost, risk, and repetitive work really sit.

  3. 03

    Decide what not to automate

    Some steps need a person. Some need fixing before anything touches them.

  4. 04

    Estimate the value

    Hours returned, errors avoided, and what it would take to get there. Estimates, with the assumptions written down.

  5. 05

    Design the simplest thing that works

    The smallest solution that fits your systems and your people.

  6. 06

    Build and integrate

    Inside the tools your team already uses, with a person in the loop wherever judgment matters.

  7. 07

    Measure, refine, and support adoption

    A system nobody uses returns nothing. Getting people to use it is part of the work.

03Engagements

Most engagements start small.

Either with finding and ranking the opportunities or with one small build. Each stage stands on its own.

  1. 01

    Assess

    Map the workflows, find where the time goes, and rank the opportunities by value and effort. You get a clear list of what’s worth doing, what isn’t, and why.

  2. 02

    Build

    Implement the highest-value opportunity first, connected to the systems you already use.

  3. 03

    Refine

    Measure what changed, adjust, and support the people using it. Ongoing advice where it’s useful.

Every estimate comes with its assumptions attached. I won’t promise a number I can’t show you the working for. Vedette, below, is that reasoning built into software.

04Pricing

Ways to start

Every price is written down before work begins. If the scope changes, you hear about it before I do anything.

Starter dashboard

The weekly report that builds itself

$750

Launch price

It goes up after my first few projects.

Still copying numbers into a weekly report? I connect one data source (Excel, Google Sheets, or your CRM) to a dashboard that updates on its own.

You get one data source, three to five KPIs, a short walkthrough so you can actually use it, and 14 days of fixes. It takes about a week. Anything beyond that gets quoted before I start.

One requirement: the data has to live somewhere a dashboard can reach it, like OneDrive, SharePoint, Google Sheets, or a CRM. A file on one person’s desktop can’t refresh itself. If your data needs cleanup first, I’ll say so and price that separately.

You pay when I hand it over and it works. If the dashboard isn’t refreshing on its own by then, you don’t pay. Before I start, I confirm your data can be connected, so I never build something that can’t work.

After-hours phone answering

Every after-hours call answered

$750

Launch price

It goes up after my first few projects.

Missing calls while you’re on a job or off the clock? I set up an AI receptionist that answers when you can’t, takes the details, and sends you a summary.

You get one phone line set up and tested with real calls, your hours and emergency rules built in, your calendar connected, and 14 days of adjustments. It takes about a week.

The service runs in your own account, so you own it. You pay the platform directly, usually $25 to $75 a month for a small service business, depending on call volume, and I’ll estimate it for yours before we start. I don’t mark it up. Where your state requires it, I add a short call-recording notice to the greeting.

Built for service businesses like contractors, cleaners, and home services.

Assess

Find out what’s worth automating

$1,000

Fixed price for one team or process area. Bigger scope gets quoted first. I map how the work happens, find where the time goes, and hand you a ranked list of what’s worth automating and what isn’t. If you build with me afterward, the fee comes off the build.

Build

Fix the one workflow that hurts most

$1,500 to $5,000

One workflow, built inside the tools you already use, with a person in the loop wherever judgment matters. Simple workflows start at the low end. Several systems or messy data push it up. You get the number, with the assumptions behind it, before you commit, and anything bigger gets quoted separately.

Refine

30% of your build fee per month

Optional, after a Build. I measure what changed, fix what breaks, and handle a small number of changes each month. It’s month to month, and you can cancel with 30 days’ notice. Many simple builds don’t need it after the first month, and I’ll tell you if yours doesn’t.

Not sure which one fits? See where your time is going and I’ll show you where I’d start.

05What I build

The build side.

Technology comes last on purpose. These are what I build once we know what’s worth building.

Automation and integration

  • Business process automation
  • AI workflow automation
  • API integrations
  • CRM automation
  • n8n workflows

AI systems

  • AI agents and agentic workflows
  • Document workflows
  • AI-assisted knowledge workflows
  • Voice agents and after-hours phone answering

Microsoft 365

  • Copilot and Microsoft 365 automation
  • Power Platform solutions
  • Dashboards and reporting

Custom software

  • Custom internal AI tools
  • Next.js and Supabase applications
06Proof

Evidence you can open.

Each of these is live. Click through and check the work.

AI opportunity assessmentLive demo

Vedette

Takes messy operational information, identifies AI opportunities, and scores them against the evidence. The model proposes, TypeScript calculates, and anything inferred is flagged as an assumption. The company in the demo is fictional.

Operations automationLive, real customers

Hawkin Standard

Intake, pricing workflow, and customer communication for a handyman business, built around how the owner actually works. A customer sends a project with photos, he sends back one price, and nothing is scheduled until they say yes. It runs real jobs today.

Complex informationLive

Pattern18

Organizes thousands of messages into timelines that stay linked to their sources, with AI-assisted pattern finding and the person using it deciding what matters. It helps people arrive organized. It does not give legal advice or replace an attorney.

TalkToHudson.ai is a safety-first conversational AI product: guardrails on every conversation, permissions over who sees what, and a path to a person when it’s unsure.

Read the build story, the TalkToHudson.ai build story
07Background

Twenty-six years in enterprise technology.

Most of it in a regulated industry, where every change passes security, privacy, and compliance review before it reaches anyone. That’s where I learned to build things people adopt, and to get them approved.

26
Years improving how work actually gets done
2,000+
Sites in the environment I lead
100+
Reports, apps, and automations shipped to real users
20+
Executive dashboards in production

I still build hands-on. Everything above, I built myself.

Have a process that’s taking too much time?

Tell me how it works today. I’ll help you work out if AI or automation makes sense, and what the simplest path forward could look like.

Prefer to talk it through? Email me.