AI implementation help
Practical help working out where AI genuinely fits in your organisation, what it will cost, and what to ignore. Independent, with no vendor commission.
Most organisations do not have an AI problem. They have a which of these forty claims is real problem, and a who is going to own this after the pilot problem.
I help with both — working out where this technology genuinely earns its cost in your specific setup, what it will take to run it after the excitement wears off, and, more often than anyone selling it will admit, where the honest answer is that it does not fit yet.
Who this suits
- Small and mid-sized organisations with a real workflow to improve, not a mandate to "adopt AI".
- Teams that have run a promising pilot and cannot work out why it will not survive contact with production.
- Anyone who has been quoted a large number by a vendor and wants an independent read before signing.
Who this does not suit
I would rather say this up front than waste a call.
- If you want someone to validate a decision that is already made, I am the wrong person and will probably annoy you.
- I do not do regulated decision-making: medical, legal, financial or safety-critical automation. Not my field, and the liability belongs with someone qualified.
- If the actual problem is that your data is a mess, AI is not the next step. That is worth hearing before you pay anyone, including me.
How it works
A first conversation, free. Thirty minutes or so. You describe the workflow, I tell you honestly whether there is anything here worth pursuing. Roughly half of these end with "not yet, and here is what would have to change first" — which is a useful outcome and costs you nothing.
A scoped review. A fixed-price piece of work with a written deliverable: what to do, what it will cost to run, what will break, what to ignore, and what to do first. Written for the person who has to sign it off, not for a conference stage.
Implementation support. Ongoing help while you build it, at an agreed day rate.
Pricing depends entirely on scope, so I quote it after the first conversation rather than publishing a number that would be wrong for most people who read it.
What you are actually buying
Independence, mostly.
I take no commission from any vendor, reseller or platform. I have nothing to sell you afterwards. If the right answer is an off-the-shelf tool at €20 a month, or a Python script somebody already wrote, or nothing at all, that is what the report will say — and that has happened often enough that it is worth putting in writing.
A worked example
Rather than list skills, here is a system I built that runs commercially every day: an AI video production pipeline that turns a Markdown file into a finished, captioned, publish-ready video. The write-up covers the architecture, the engineering decisions, the measured per-video cost, and the parts that are fragile.
The most useful finding from it: human handling was 89% of the cost per video, and every AI service combined was under 6%. Most organisations plan the opposite way, because API pricing is the number with a label on it.
The written record of how I think about this is that case study, the blog, the videos and the tools. Read some of it first and decide whether you want this particular brain on your problem. That is a better filter than anything I could claim about myself on this page.
Get in touch
Email zeroesonesandyou@gmail.com with a couple of paragraphs about the workflow and what you have tried. That is enough for me to tell you whether it is worth a call.