Imagination first
Building AI skills, agents and CLI (MCP tools) across Claude, Codex and Gemini. A deterministic scaffold of strong Python and Object Oriented Design (OOD) underneath, model judgement on top.
“Don't chase the models. Build the scaffold they rise into. Towards imagination-first AI.”
Co-founder of Wordtracker
I co-founded Wordtracker in 1998, the first keyword tool, the one that kickstarted SEO.
It's still going 28 years on. I led its technology from two developers to around 25 and scaled turnover roughly 100× over a decade, never stepping away from the codebase.
I've been a Python developer for a decade.
Agentic engineer and founder-operator. A repeatable software factory underneath, with AI models supplying the judgement—not the structure.
Since January I've gone AI-first, not a line hand-written since. Everything here is what that's produced: skills, agents and CLI (MCP tools) on Claude, and a methodology, Stride, for shipping AI-built software that lasts. I move fast because I iterate in the open. Remote-native, self-directed, available now.
A constant for nearly two decades
Wordtracker has been that constant for me — dependable, insightful, and backed by a team that truly understands the needs of digital marketers.
Reliable since 2006
Over the years, I've relied on many tools in the industry, but few have demonstrated the same level of reliability, clarity, and long-term value as Wordtracker.
Exceptional team support
The relationship and level of support I have received from Mike Mindel and his team have been exceptional. Their professionalism, responsiveness, and genuine commitment to their users stand out in an industry where that level of service is not always guaranteed. It's clear that they care not just about the product, but about the success of the people using it.
The most relevant keywords without the complexity
Your tool is truly amazing because your engineers figured out how to get the most relevant keywords for each topic without all the unnecessary complexity that causes distraction.
Useful for marketing and academic research
I used your tool during my bachelor's and master's degrees. It's useful not only for digital marketing but also for educational research.
Over a decade — and it always delivers
I have been using your set of tools for over a decade because it always delivers. I am especially grateful for how you maintain the tool and constantly serve the most relevant keywords in a way that is ready to be implemented in my evolving research.
Keyword research made easy with Wordtracker
The additional feature of IAAT makes the keyword research easy along with PPC research.
Less expensive than Ahrefs for basic SEO
You rock and look forward to using you guys for the foreseeable future. Thanks for being less expensive than Ahrefs for my basic SEO needs.
Great tool for SEO specialists
Wordtracker is a great tool that helps with all daily SEO tasks. The best part is its free version which allows users to get started with no cost.
Best tool for beginners in Search Field
Wordtracker offers a refined user interface, and even beginners like interns can start doing extensive keyword research.
Stands the test of time
This tool has had nothing but a positive impact on our business and helps us tend to our clients everyday. Our entire SEO department uses Wordtracker daily and we find it an indispensable tool — selected because of its dependability over time and results.
Invaluable for content marketing
Wordtracker has been invaluable for my company as a content marketing research tool. It's the best keyword analysis tool because it's easy to understand, use, and find the information you're looking for — and the most well-known and trusted amongst its competitors.
Good bang for the buck
Good bang for the buck. Makes the workflow easier by being able to extract data faster, quick for my team to learn how to use, and lots of data — really, really lots of data. It gives you data which Google used to give you.
Founded and grew Wordtracker into an internationally recognised keyword research platform serving customers worldwide over 28 years.
Financial Times
“Brothers’ fast-track to success is being hot in the pursuit of words…”
Read article →
Mike is a pioneering figure in search marketing — CEO and co-founder of Wordtracker, one of the internet's first commercial keyword research tools. Wordtracker put keyword research at the centre of digital marketing and helped kickstart SEO as we know it.
Pioneering keyword research — Wordtracker let SEOs find profitable niches and see exactly how often people searched for specific terms, shifting SEO from guesswork to strategy.
Education and evangelism — gave many talks at industry conferences on technical SEO, content and search algorithms.
Industry influence — through the 2000s and 2010s, collaborated with SEO pioneers such as Aaron Wall of SEO Book and Ken McCarthy of The System Seminar to shape best practices still used today.
The AI and agent stack leads, because that's where the work is now. The rest is the ground it stands on.
Skills, agents and CLI (MCP tools) on Claude.
Core languages I use daily.
Python web frameworks.
Modern React applications.
Large-scale data pipelines.
Provisioning and deployment.
Confidence before shipping.
Design for change.
Turning ambiguous ideas into products.
How I work.
Orchestration is where I'm pushing next: wiring agents and pipelines into systems that run themselves.
Don't chase the models. Build a stable scaffold they rise into. That scaffold is a harness: it blends deterministic craft (Python, AST analysis, skills written in markdown) with probabilistic model judgement, the best of both worlds working together. As my context engineering sharpens and the models improve underneath it, the two compound. A rising tide that lifts both.
Here's a concrete one. I built a /restructure skill this year. It reorganises code that's grown too big and broken the Single Responsibility Principle. Instead of the model reading a whole codebase and guessing where to act, Python scans the file system and scores every file for "tension": AST analysis for size and coupling, git history for churn, a test-coverage safety multiplier. The model gets hard data on where the tension actually sits, complexity times impact, and spends its judgement on the restructuring, not on hunting for the problem. Deterministic code finds the facts. The model does the thinking. I've built a library of patterns like it since.
Another: lexicographic rules beat pattern taxonomies for classification. A lexicographic rule orders cues by importance and acts on the first one that discriminates, like a dictionary comparing the first letter and only checking the second if those tie. Far more robust than pulling a set of patterns out of the data and forcing everything new into them. I'd assumed AI would do best fed a ready-made pattern library. Turns out it works the patterns out itself, given a simple lexicographic steer.
The scaffold rests on plain deterministic craft. It's the one line I keep coming back to:
When faced with two or more alternatives that deliver roughly the same value, take the path that makes future change easier.
— Andrew Hunt & David Thomas, The Pragmatic Programmer
My take on Sandi Metz's 99 Bottles OO exercise puts that to work: polymorphism instead of conditionals, methods of five lines or fewer, each class with a single reason to change. The 0, 1 and 6 cases become subclasses that override only what differs. It's the same discipline the /restructure skill automates.
class BottleNumber:
@classmethod
def from_number(cls, number):
return {
0: BottleNumber0(number),
1: BottleNumber1(number),
6: BottleNumber6(number),
}.get(number, BottleNumber(number))
def __init__(self, number):
self.number = number
def quantity(self):
return str(self.number)
def container(self):
return "bottles"
def successor(self):
return BottleNumber.from_number(self.number - 1)
class BottleNumber0(BottleNumber):
def quantity(self):
return "no more"
def successor(self):
return BottleNumber.from_number(99)
More on GitHub:
Built Stride, a way to run a dev team on a Kanban board where the team is AI. Every card, every change, visible and accountable. It gives AI-assisted development guard rails (traceable, reversible, tied to a Vision) so teams ship AI-built software that lasts past day 30, instead of vibe-coding software that collapses under its own debt.
It runs on Claude Code with Linear / Kanban delivery, atomic commits behind pull requests. Bounded pieces go to agents, and every change stays reviewable through the board.
Presented it in a 25-minute talk, "Stop building on quicksand — ship AI apps that last past day 30", at the AI Global Festival (Suffolk, 2026).
The talk argued against vibe coding and for staged agentic engineering: start with a shared Vision, telegraph work across the team, give agents bounded pieces to deliver, then keep every change reviewable through the board, commits and pull requests.
The same methodology delivered the smart-glasses spike below. It's the method in live use on paid client work, not just theory, and I haven't hand-written a line of code since January.
For an early-stage AI startup, ran a fixed-scope two-day go/no-go spike proving a Convex reactive-database web app could drive Even Realities G2 smart glasses with real-time two-way sync.
Built a test Convex app and demonstrated a clean round trip between the phone's companion app and the glasses lens. Because the client's platform runs on the same reactive database, the result carried straight to it.
Delivered remotely and asynchronously with a shared repo and a written findings note. (Client named on request, subject to NDA.)
The keyword-research / search-intelligence tool I co-founded with my brother in 1998 and have led the technology for ever since. Financial Times–profiled; scaled ~100× over its first decade and ran profitably at national-press scale.
Co-founder & developer, 1998–2013: the technical half of a two-developer team; the CTO role grew as the team grew, to around 25 at peak.
CEO / CTO (dual-hat), Mar 2013 – Jan 2026: ran the company and owned the technology at once.
Director of the family photographic business in Mill Hill, London, selling photographic albums and storage pages.
| Year | Title | Where |
|---|---|---|
| November 2008 | Five Powerful Ways to Use Keywords | Ken McCarthy's System Seminar |
| June 2017 | Get Noticed! (The Art of Getting Attention) | Spoken-word performance |
| November 2023 | Spoken word artist | The Red Lion, Eynsham |
| December 2023 | Spoken word artist | The Art House, Southampton |
| 2026 | Stop building on quicksand — ship AI apps that last past day 30 | Talk — AI Global Festival, Suffolk |
| Years | Institution | Study |
|---|---|---|
| 1993–1996 | Reading University | BA (Hons) Film & Drama, 2:1. |
| 1991–1992 | Bristol University | Computer Science & Maths (two terms). |
| 1986–1991 | Mill Hill School | A-levels in Economics, Maths, Physics and English Literature; 12 GCSEs, including English, Physics and Computer Studies at grade A. |
Same instinct, different medium. Teaching a room to land a joke or an agent to land a refactor comes down to the same thing: a clear method, fast feedback, practice until it sticks.
Ran the Wam Bam Club for 10 years — a West End burlesque cabaret at The Café de Paris (Leicester Square) and The Bloomsbury Ballroom (Holborn); comedy, cabaret and controlled chaos, engineered to leave the room buzzing (TripAdvisor reviews).
Don't Let Him In (2011) — Horror feature film. A nerve‑bludgeoning, slow‑burning psychological horror/slasher in which a rural weekend getaway turns into a deadly double date when a charming stranger may be the serial killer known as “The Tree Surgeon.” Released internationally (incl. US Jan 2012; UK Oct 2012) and sold in major retail (incl. Walmart).
A filmmaker too — Burning Man | Release! (2015).
Across these projects I've worked as a film editor — now specialising in AI, with tools like Seedance 2.5 and Kling — as well as camera operator, director of photography and producer.
Wow. Watching your film now. The music and imagery is SUBLIME brother. Wow. Mediated ecstasy. Love it.
— Jason Silva, Shots Of Awe
Also created and ran comedy bootcamps in partnership with award-winning comedian Joe Bor.
1 Line Wonders (Comedy Challenge)
A one-session taster: complete beginners learn to write original one-liners, workshop them in small groups, then perform live on stage.
5 Minute Wonders
A multi-week programme taking novices from zero material to a confident five-minute stand-up set, building material and stagecraft week by week, ending in a live showcase.
I think 5 Minute Wonders was productive, welcoming, collaborative and inspiring! I bet that doesn’t happen too much on the comedy circuit!!
— Judy Goldberg
It's the same teach-through-practice pattern that runs through Stride, just a different medium. A method, fast feedback, and reps until it sticks. It's how I design skills for agents too.
Dan Shapiro maps AI-assisted development onto a ladder, from spicy autocomplete to the software factory. It's the self-driving-car analogy for code: at the bottom you grip the wheel, at the top you ride in the back and just set the destination.
| Level | Stage | What it means |
|---|---|---|
| 0 | Spicy autocomplete | The AI serves as a reference tool or enhanced search, like a smarter Stack Overflow. The developer manually writes all the code, using AI for specific snippets or answers. |
| 1 | Coding intern | The AI writes unimportant or boilerplate code. The developer prompts for specific functions but immediately reviews and integrates the output. |
| 2 | Junior developer | An interactive "pair programmer" partnership. The developer and AI trade off control, with the human reviewing code as it's generated in real-time. |
| 3 | Developer | The AI generates the majority of the codebase. The developer reviews everything the AI does, acting as the bottleneck for verification before progressing. |
| 4 | Senior developer | The AI runs unattended for long periods, handling complex tasks. The human trusts the system's self-checks, only checking the final results much later. |
| 5 | Software factory | The engineer manages the goals and the system, not the code. They provide plain English descriptions. The AI defines implementation, writes code, tests, fixes bugs, and ships. |
Most teams sit around level 2 or 3, a human as the bottleneck on every change. I work at level 4 now, building towards 5.
The real goal isn't running the codebase at 4 and 5. It's running the business there. Skills, agents and CLI (MCP tools) that own the work end to end, people setting the goals, the system checking itself. A lights-out factory for the busywork, so the team spends its time on what only people are good at.
Levels 4 and 5 mean the system checks its own work and adapts. If we're reviewing code for QA, we're not there yet.
My model for it comes from military situational awareness. See where you are, adapt, take the next step towards the goal, repeat. That's the direction I build in.
Here's the question I keep asking: how do you build a business that creates more value as a side effect of the models getting stronger? Get that right and every release works for you, not against you.
My answer is the rising tide. The markdown skills a team builds are a collaboration between agent and human, and they sharpen with every pass. Each pass improves the question. A better question and a better model compound, so as the models rise the same skills throw off more value, for free.
The flip side is a warning. If there's no moat in the business, AI will consume it all. Anything that's pure AI has no future, because a stronger model just does it cheaper next quarter.
So build on what AI can't own. Networks are the strongest moat there is: humans connecting with each other. AI can augment that connection, make it richer and faster, but it can't replace the people. That's the ground worth building on.
I run a second brain. It's the Tiago Forte idea: capture what you learn, get it back when you need it, and action it now. I can pull a detail from a conversation months old and use it today.
That's context management, the same skill the agent work rests on. Wherever I'm working, it's the edge. Work gets done better and faster, with the best knowledge to hand. And I don't keep it to myself. I coach people to set up their own, so this AI-powered way of working spreads through the company and lifts everyone it touches.
I run it on Cmux and Obsidian. Cmux holds several second brains at once. Obsidian's where I read them. So I've got multiple knowledge bases, all talking to each other.
So the right knowledge lands just in time, for the exact job in front of me.
I'm AI-native too. I'll run several Claude, Codex and Gemini CLI jobs at once, not one at a time.
For the hard calls I lean on more than one model. OpenRouter's Fusion fires a prompt at a panel of models in parallel, with web search. A judge weighs the answers, and my model writes a better one off the back of it.
I do the same with documents. Fan one out to several frontier models and pull back a synthesised critique, a build on Karpathy's llm-council.
Three spaces I'm involved in and would like to explore further.
| Industry | Focus |
|---|---|
| Trading | Commodities, options, crypto. Markets run on data, and I like building the tools that read it. |
| Marketing | Landing pages, advertising, media buying. I built my own MDX system, lander, to keep content and design apart, so a page stays quick to change. |
| Search, AEO and GEO | The same ground I've worked for 28 years, now shifting to AI answers. I know how people find things, and how that's changing. |
| Work | The work |
|---|---|
| Agentic systems | Building skills, agents and CLI (MCP tools) so a team can focus on what only people are good at. |
| Data analysis | Crunch and analyse large volumes of data to detect patterns (Redshift, BigQuery). |
| Trading | Writing trading systems. |
| APIs | Writing APIs (FastAPI / Pydantic) which serve valuable data. |
| Creative problems | Solving interesting creative problems, like the two-way sync between Convex and the Even Realities glasses. |
| Term | Detail |
|---|---|
| Availability | Immediately, remote — UK-based. |
| Engagement | Contract or employment, equally welcome. Ideal is part-time — 2 days a week, 3 at most — with room to scale up for the right project. |
| How I work | Remote-native and self-directed. Happy to lead a team of coders and agents, or embed in yours. |
| Experience | 25+ years shipping production software; Wordtracker co-founder. |
| Invoicing | Through Webventurer Ltd. |
I'm also available for part-time or full-time employment, as well as contract work.
| Term | Detail |
|---|---|
| Availability | Immediately. |
| Type | Part-time or full-time. |
| Location | Remote, hybrid or on-site — UK-based. |
Worth a conversation? I'll give you the live tour: Stride and the tension-score script. Available now for contract work through Webventurer Ltd, or for part-time or full-time employment.