An AI app developer who builds the pipelines
not just the prompts.
Six months of working AI applications. One first-page Google ranking against agencies and recruiters. Custom pipelines for SEO, content, sales, prospecting, estimating, legal workflows, and finance. The apps are private. They are applied for you, through a consultation.
6+
Months of working AI pipelines
22
Governed articles in one run
1st
Page Google ranking on a competitive query
SERPs + LLMs
Built for both ranking surfaces
Generic AI content does not rank. This is what does.
Pipelines, not prompts
Each app is software. Stages, evaluation gates, governance, source-of-truth handling, deterministic checkpoints, and human review. Six months of engineering, not a weekend of prompts. The output behaves predictably because the system was built to.
No SaaS, no licence
The apps are private. There is no public dashboard, no signup, no plan tier. The value is in the operator applying the system, not in shipping the software to anyone with a credit card. You engage me. The system is applied to your pages. That is the model.
Built for SERPs and LLMs
Pages are engineered to rank in Google search and to be cited in ChatGPT, Claude, Perplexity, and Gemini answers. Most AI SEO providers cannot speak to LLM ranking yet. The toolbox treats both surfaces as one engineering problem.
Operator with proof
I ranked my own service page on the first page of Google for a competitive Cape Town query, against recruiters, agencies, and large software companies. The same recipe, the same pipelines, applied to your business pages.
Private AI applications, accessed through consultation
SEO AI Toolbox
A multi-pipeline system applied to one ranking goal at a time. Combines evidence gathering, search-opportunity research, governed content production, and downstream amplification into one engineering process. The system used to produce first-page Google rankings on competitive service queries.
Typical work: Live application page, read what it does and why it is not for sale.
See the SEO AI Toolbox›Site Intelligence
Crawls and captures the full current state of a website: HTML, screenshots, headings, links, sitemap, technical observations. The output is the source material every downstream decision is anchored to.
Typical work: Available through consultation.
Content Marketer
Produces drafts under governance gates: no invented facts, no AI-tell phrasing, every article links a service page and a case study. 22-article governed runs already on the books.
Typical work: Available through consultation.
Prospect Finder
AI prospect discovery and qualification for service businesses that want better-fit leads instead of bulk lists. Inclusion and exclusion criteria, qualification scoring, evidence URLs.
Typical work: Available through consultation.
Sales Team
AI sales workflow support for lead memory, qualification, follow-up, sales notes, and pipeline tracking. Designed for senior operators, not call centres.
Typical work: Available through consultation.
Estimator, Marketing Planner, Financial, Legal
Additional AI apps for estimation, marketing planning, financial assessment workflows, and legal case-assessment workflows. Each tackles a discrete business workflow with its own outputs.
Typical work: Available through consultation. New apps added as the portfolio grows.
See the full AI app portfolio›Why you cannot get the same outcome cheaper or faster elsewhere
No SaaS, no licence, no reseller
There is no signup, no plan, no API key for sale. The pipelines are owned and operated. You engage the operator. That is the only way to access the system.
Six months of iteration is the moat
The eval gates, governance, source-of-truth handling, and stage discipline are months of work. Reproducing them is a real engineering project, not a prompt template.
Selling it would commodify it
A self-serve version would lose the gates that prevent generic AI output. The output is good because a senior operator applies the system, not despite it.
One operator, limited engagements
No account managers, no handoffs, no monthly content treadmill. Limited project slots per quarter. Fit is decided in the consultation.
Proof you can verify in 30 seconds
/software-developer-cape-town, first page of Google
Live page on this site that outranks recruiters, agencies, and large software companies for the query "custom software developer Cape Town". Open it. Inspect the schema in DevTools. Look at the sitemap entry. The system that produced it is the system on this page.
Open the live page›22-article governed content run
A single run produced 22 drafts, ~45,000 words. Every article passed governance checks. Every article links to a service page and a case study. No invented facts. No AI-tell phrasing. Recorded project logs.
Site Intelligence scan output
A custom site scanner captures full HTML, screenshots, metadata, headings, links, sitemap, response codes, and load times. The output is the source material every downstream decision is anchored to, before any opinion is written.
Cape Town ranking case study
Full walkthrough of how the page was built, the recipe, the pipelines, the schema, the internal links. Written as an engineering case study, not a marketing post.
Read the case study›Why these systems exist
Twenty-five years in technology. Fifteen years building production software for the kind of work that has to actually run: e-commerce systems for WooCommerce and Shopify shops across the US, UK, Europe, Australia, and South Africa; a POPIA-compliant medical and dental practice platform with patient records, clinical notes, and medical aid claims; and BX1X, my own SaaS business operations platform with 37+ modules covering billing, e-commerce, inventory, bookings, and CRM. Systems that ship, run, and hold up to audits.
Running my own SA-registered company since 2001 means I also see software from the business-owner side: scopes, quotes, deadlines, support tickets, cash flow, content that has to actually drive enquiries. The pipelines on this site were not designed in a vacuum. They exist because the same problems kept showing up in my own business, and the available off-the-shelf AI tools did not survive contact with real production work.
Three years of building with AI systems day-to-day, and watching them produce confident output that turns out to be wrong. The strengths and the failure modes of large language models are concrete to me, not theoretical. That is why these pipelines have eval gates, source-of-truth handling, and governance rules. I built these apps first for my own business. After watching them produce real results, including the first-page Google ranking that prompted this site, I decided to apply the same systems for clients. The apps came from doing the work, not from a productisation strategy.
25
Years in technology
15+
Years production software
3+
Years applied AI engineering
2001
Running own SA company since
SERPs and LLM answers, one engineering problem.
Most AI SEO providers cannot speak to LLM ranking yet. Pages are engineered with strong entity definitions, paragraph-level facts, FAQ schema, citable URLs, and consistent identity across pages. These are the structures that LLM crawlers reward. Built to be cited in ChatGPT, Claude, Perplexity, and Gemini, not just to rank in Google.
How It Works
Book the consultation
Tell me what you want to rank for, in Google or in LLM answers. One field on the form, one short message.
Fit is decided on the call
We talk through the buyer, the query, and the proof you already have. If it is a fit, we move on.
The system is applied
Search opportunity map → page plan → content cluster → internal link plan → schema and indexing → measurement. Each stage is a deliverable.
Typical Engagements
This is not a monthly content retainer. It is a search-growth system built in stages, with proof at every stage, scoped to one ranking goal at a time.
- Search opportunity map: identify the queries the business can realistically win, in Google and in LLM answers.
- Page plan and content cluster: the page that will carry the ranking, plus the support content that authority-feeds it.
- Internal link plan: the authority graph that turns pages into a topical cluster instead of orphans.
- Schema and technical SEO: ProfessionalService, LocalBusiness, FAQPage, BreadcrumbList. These are the structures Google and LLMs read.
- Measurement: indexing, impressions, clicks, and LLM citation tracking over the first 90 days.
One operator. The whole engagement.
No account managers. No handoffs. You work directly with me from the first conversation to the final measurement review.
Frequently Asked Questions
Is this just ChatGPT in a wrapper?
No. Each app is a custom pipeline with stages, governance gates, deterministic checkpoints, and human review. Output is engineered, not generated. Six months of iteration separates this from prompt engineering.
What does an AI App Pipeline Developer actually do?
Designs, builds, and operates AI pipelines as software. The role combines software engineering, system design, and applied AI, not just prompt writing. The pipelines are real, with stages, evals, gates, and recorded project logs.
Why are the apps not for sale?
The value is in the operator applying the system, not in software being run by anyone with a credit card. There is no SaaS plan, no licence, no public API. You engage me. The system is applied to your pages.
Can our pages actually appear in ChatGPT or Perplexity answers?
Pages are engineered for the structures LLM crawlers reward, strong entity definitions, paragraph-level facts, FAQ schema, citable URLs, and consistent identity across pages. We do not promise specific citations. We build pages designed to be cited and we track results.
How is this different from an SEO agency or a content writer?
An agency sells templates and retainers. A content writer sells words. This sells a search-growth system built in stages, scoped to one ranking goal at a time. One senior operator, no account managers, no content treadmill.
Do you guarantee a first-page Google ranking?
No one can responsibly guarantee a Google ranking. The system is built to compete for first-page search visibility and has already produced a first-page result on a competitive Cape Town service query. The aim is proof-led work with measurable outcomes.
What do you need from us to start?
A short description of the business, the customer you want to reach, and the query or problem you want to win. One short message on the consultation form. The rest happens on the call.
What markets do you serve?
Based in Cape Town, South Africa. Currently working with clients in South Africa and internationally. The system is not market-specific; it is intent-specific.
Tell me what you are trying to rank for.
The contact form has one optional field for the query you want to win. I review every enquiry personally. If we are not a fit, I will say so on the call.
Looking for a senior AI app developer who can prove the system works?
I review every enquiry personally. Engagements are scoped after the consultation, never before.