One of the most difficult challenges in human performance research is comparing results across different populations.

Fitness data is collected everywhere. Schools, sports programs, military organizations, health systems, wearable devices, and fitness applications all generate enormous amounts of information. Yet meaningful comparison remains surprisingly difficult because the underlying measurements are often different.

Different tests measure different things. Different organizations use different standards. Different countries collect different types of data. Even when two groups appear to be measuring the same concept, the methods used may not be comparable.

As a result, many discussions about population fitness rely on assumptions rather than direct comparison.

Cross-Population Comparison is one of the core objectives of the Global Fast Fit project.

The idea is straightforward: if individuals from different populations are evaluated using the same benchmark, under the same rules, with the same verification requirements, meaningful comparison becomes possible.

This principle was one of the primary reasons the GFF Standard was developed. Rather than relying on country-specific testing protocols or organization-specific fitness assessments, participants are evaluated using a common benchmark that can be performed across diverse environments. The goal is not to eliminate differences between populations, but to create a common reference point through which those differences can be studied.

Cross-population comparison is not limited to geography. The same methodology can be applied to age groups, genders, occupations, training backgrounds, and other population segments. A benchmark becomes more valuable when it can support comparison across multiple dimensions rather than within a single group.

This capability has become increasingly important as the Global Fast Fit dataset has expanded. Thousands of benchmark submissions and exercise records collected across multiple countries have created opportunities to examine performance patterns that would be difficult to observe within smaller or isolated datasets. Questions about how fitness varies by age, training history, location, or demographic group become easier to explore when all participants are measured against a common standard.

The broader Human Performance Intelligence (HPI) initiative was built in part to support this type of analysis. While individual benchmark results provide value on their own, larger datasets make it possible to study trends, distributions, and relationships across populations. Cross-population comparison transforms isolated fitness tests into a growing body of comparative human performance data.

The objective is not to declare one population stronger, healthier, or more capable than another. Human performance is complex and influenced by many factors. Instead, the goal is to provide a framework through which meaningful differences can be measured, studied, and understood.

In many fields, progress begins with the ability to compare. The same is true for human performance. Without common benchmarks, comparisons become difficult. Without comparison, patterns remain hidden.

Cross-Population Comparison is therefore more than a statistical exercise. It is one of the foundational reasons global benchmarks exist in the first place. By establishing a common standard and a common methodology, Global Fast Fit seeks to make meaningful comparison possible across the populations that make up the world.

Most fitness systems measure only a small portion of human performance. A running test may measure cardiovascular capacity. A strength test may measure muscular performance. A wearable device may measure activity levels. A health assessment may focus on biomarkers. Each provides useful information, but each captures only a fragment of the larger picture.

Human Performance Index (HPI) was developed to address this challenge.

HPI is an effort to create a more comprehensive view of human performance by integrating multiple forms of evidence into a single framework. Rather than focusing on one exercise, one test, or one device, HPI seeks to understand how different measurements relate to one another and what they collectively reveal about an individual's capabilities.

The idea emerged from a simple observation: human performance is multidimensional. A person may possess excellent cardiovascular fitness but limited strength. Another may demonstrate impressive strength while struggling with mobility or endurance. Looking at a single metric often produces an incomplete understanding of overall performance.

The Global Fast Fit benchmark became one of the foundational components of HPI because it provides a standardized, verifiable measure of functional fitness. However, HPI extends beyond GFF alone. It is designed to incorporate additional forms of performance evidence, including exercise records, movement assessments, activity data, and other measurable indicators of physical capability.

A key objective of HPI is creating comparability. Human performance data is often fragmented across devices, applications, fitness programs, and health systems. Measurements may be collected using different standards, making meaningful comparison difficult. HPI seeks to provide a framework through which diverse performance data can be evaluated within a common structure.

Verification also plays an important role. Many performance systems rely heavily on self-reported information or proprietary scoring methods that are difficult to examine independently. HPI places greater emphasis on observable, measurable, and verifiable performance whenever possible. The goal is not simply to generate scores, but to create confidence in what those scores represent.

The broader significance of HPI extends beyond individual fitness assessment. As larger datasets are collected and standardized, opportunities emerge to study patterns across populations, age groups, training methods, and environments. Questions that are difficult to answer using isolated records become more accessible when performance data can be evaluated at scale.

This is one reason HPI is closely connected to the larger Global Fast Fit ecosystem. The benchmark provides a common reference point, while the surrounding data infrastructure makes it possible to analyze performance across thousands of observations rather than isolated individual results.

Human Performance Index is ultimately an attempt to move beyond isolated fitness metrics toward a more integrated understanding of human capability. It recognizes that performance is complex, that meaningful measurement requires multiple perspectives, and that better data creates opportunities for better insights.

As the collection of human performance data continues to expand, HPI represents an effort to transform individual measurements into a broader system of knowledge—one capable of helping researchers, coaches, organizations, and individuals better understand how people perform, improve, and age over time.

The Global Fast Fit Pro benchmark was the original version of the GFF system.

The benchmark consists of four components:

• 30 Pushups

• 30 Plank Leg Lifts

• 30 Squats

• 500 Meter Run

While GFF Standard was ultimately adopted as the flagship benchmark for broad population use, GFF Pro remains an important part of the Global Fast Fit ecosystem.

The relationship between the two benchmarks reveals one of the most important lessons learned during the development of GFF.

A benchmark is only useful if people can actually perform it.

Early testing showed that GFF Pro was highly effective at differentiating fitness levels among active individuals. However, it also revealed that many participants who considered themselves reasonably fit were unable to complete the benchmark successfully. Pushups proved to be a particularly significant barrier, especially among older adults, women, and individuals with limited training backgrounds.

This finding was important because the long-term objective of Global Fast Fit was not simply to identify elite performers. The goal was to create a benchmark that could be deployed globally and serve as a common reference point across diverse populations.

Rather than abandoning GFF Pro, the project evolved into a two-tier system.

The GFF Standard routine became the primary benchmark for broad population assessment and comparison. GFF Pro remained available as a more demanding benchmark for individuals seeking a higher standard of performance.

This distinction allows the system to evaluate functional fitness at different levels while maintaining consistency in methodology and verification.

Like the Standard benchmark, GFF Pro is designed around observable physical performance rather than self-reported fitness. Participants demonstrate their abilities through a structured sequence of exercises that can be reviewed and verified through video evidence. This emphasis on verification is one of the characteristics that distinguishes GFF from many fitness assessments that rely heavily on estimates, questionnaires, or indirect measurements.

The existence of GFF Pro also provided valuable information about global fitness itself. One of the most significant outcomes of the project was not the benchmark design, but the insight gained from observing how different populations performed when asked to meet a consistent standard. The gap between perceived fitness and demonstrated fitness was often larger than expected.

Today, GFF Pro serves several purposes. It provides a more challenging benchmark for highly motivated participants, creates additional differentiation among stronger performers, and continues to contribute valuable data to the broader Global Fast Fit research and benchmarking effort.

The development of GFF Pro ultimately influenced the creation of the GFF Standard routine, making it one of the most important components in the history of the project. Without Pro, the team would not have discovered where the balance between rigor and accessibility truly existed.

In that sense, GFF Pro is more than a harder benchmark. It is the benchmark that helped define what Global Fast Fit would become.

The Global Fast Fit Standard is the flagship benchmark of the Global Fast Fit system.

It was designed to answer a simple question: 'Can a person demonstrate a practical baseline level of functional fitness using a small number of exercises that can be performed almost anywhere in the world?'

Many fitness assessments require specialized equipment, laboratory testing, expensive devices, or lengthy protocols. While those approaches can provide useful information, they are often difficult to deploy at scale.

The GFF Standard was built around a different goal: creating a benchmark that is simple enough to be performed in diverse environments while still measuring multiple aspects of physical capability.

The benchmark consists of four components:

• 15 Pushups

• 15 Plank Leg Lifts

• 15 Squats

• 250 Meter Run

Together, these exercises evaluate upper-body strength, core stability, lower-body function, and cardiovascular capacity. Rather than measuring a single fitness attribute, GFF Standard is intended to assess an individuals level of functional capability across several major categories of movement.

The choice of exercises was not accidental. One of the challenges in building a global benchmark is balancing accessibility with difficulty. A benchmark that is too easy provides little information. A benchmark that is too difficult excludes a large portion of the population.

The GFF Standard routine was developed after extensive testing and comparison against alternative approaches. Early versions of the benchmark were significantly more demanding. While those versions provided greater differentiation among highly fit individuals, they proved impractical for broad population deployment. The Standard emerged as a compromise between rigor and accessibility, allowing meaningful comparisons across age groups, genders, countries, and fitness backgrounds.

Another important feature of GFF Standard is verification. Unlike self-reported fitness questionnaires or estimated scores, GFF is designed to support video-based validation. Participants can submit evidence of performance, allowing results to be reviewed and verified. This creates a stronger foundation for comparison than systems that rely solely on self-reported data.

Over time, GFF Standard has become more than a fitness test. It serves as a common reference point within the broader Global Fast Fit ecosystem. Results can be compared across populations, used in longitudinal tracking, incorporated into health and performance studies, and connected with other human performance data.

The value of GFF Standard is not that it perfectly measures every aspect of fitness. No single benchmark can do that. Its value lies in providing a practical, repeatable, and globally deployable reference point that allows people from different backgrounds and locations to be measured against the same standard.

In a world where fitness is often assessed using incompatible methods, the Global Fast Fit Standard was created to provide a common language for functional fitness.

The Global Fast Fit benchmark did not begin as an attempt to create another fitness challenge.

It began with a much larger question:

Is it possible to create a practical fitness benchmark that can be used across countries, populations, and environments while still producing meaningful, verifiable results?

At the time, the fitness landscape was already crowded with tests, assessments, scoring systems, and performance standards. Some focused on strength. Others emphasized endurance, body composition, athletic performance, or health outcomes. Many required specialized equipment, laboratory testing, trained personnel, or facilities that were not universally available.

What was missing was a benchmark that could be deployed broadly, verified consistently, and compared across diverse populations.

The earliest versions of Global Fast Fit focused on functional movement and observable performance. Rather than relying on questionnaires, estimates, or indirect measurements, the goal was to create a benchmark based on tasks that people could actually perform and demonstrate. Every exercise had to satisfy several requirements. It needed to measure a meaningful aspect of physical capability, require little or no equipment, be practical in different environments, and be suitable for video verification.

This process led to the development of what would later become GFF Pro.

The original benchmark was intentionally demanding. The objective was to establish a standard that represented a meaningful level of functional fitness rather than a minimal participation threshold. Early testing demonstrated that the benchmark successfully differentiated performance levels among active individuals and athletes. However, it also revealed something unexpected.

Many people who considered themselves reasonably fit could not complete the benchmark.

Pushups emerged as a particularly significant obstacle. Across different age groups, populations, and training backgrounds, performance dropped more quickly than anticipated. What initially appeared to be a reasonable standard proved substantially more difficult for the general population than expected.

This discovery became one of the most important findings of the project.

The challenge was no longer designing a rigorous benchmark. The challenge was designing a benchmark that could function globally.

A benchmark that only a small percentage of the population can complete may be useful for evaluating high performers, but it is less useful as a global reference point. The project therefore shifted from creating a single benchmark to creating a benchmark system.

GFF Pro remained as the higher-level standard, while a second benchmark was developed to support broader participation. This eventually became the GFF Standard.

The Standard preserved the core philosophy of the project while making the benchmark accessible to a much larger percentage of the population. The objective was not to make the test easy. It was to establish a baseline level of functional fitness that could be applied consistently across different countries, age groups, and backgrounds.

As deployment expanded, the benchmark design continued to evolve. Video verification became a central component of the system, helping create a stronger evidentiary foundation than self-reported performance alone. Data collection efforts expanded across multiple countries, creating opportunities to observe how the benchmark performed in different environments and populations.

These experiences influenced the development of the broader Global Benchmarking Methodology and eventually contributed to the creation of Human Performance Intelligence (HPI). What began as a fitness benchmark gradually became part of a larger effort to understand human performance through standardized, verifiable measurement.

Perhaps the most important lesson from the benchmark design process was that accessibility and rigor are not opposing goals. A successful benchmark must balance both. Too easy, and it provides little information. Too difficult, and it excludes much of the population it is intended to measure.

The history of the Global Fast Fit benchmark is therefore not simply the history of a fitness test. It is the history of an ongoing effort to create a common reference point for human performance—one capable of supporting meaningful comparison across populations, countries, and generations.

The benchmark that exists today is the result of years of testing, refinement, deployment, and observation. More importantly, it is the result of learning what happens when a benchmark is subjected not just to theory, but to the realities of the real world.

How twelve women over thirty in Nyandarua are climbing a dynasty of African queens — one skip at a time.

It started with a walk

Early this year, in our capacity as Nyandarua County's official fitness partner, we were asked to help condition Rufas — the county's immunization champion — for a 62-kilometre walk to get people talking about immunization. He made the distance. And afterwards, in the speeches, someone handed us a challenge that wasn't quite a contract and wasn't quite a joke: make Nyandarua the fittest county in the republic.

That was the genesis of everything that followed.

The first question wasn't what exercise — it was for whom. Here a doctor's eye had to step in. Our outpatient clinics are full of people carrying lifestyle diseases that movement could have softened, or prevented entirely. And one group keeps getting squeezed out of the conversation: women over thirty. This is the season where you're raising children, holding down work, and running a home all at once — and exercise is the first thing to fall off the list. Not because these women don't care, but because there's no hour left to give. So we built for them, on purpose: the hardest group to reach, chosen deliberately. We called it the Nyandarua Women 30+ Movement Series.

Then — why rope? Because it costs almost nothing, it can be done in your own yard, and, this part matters, it's already ours. Most women in Nyandarua grew up skipping. I was born and raised here, so I can say that with confidence. We weren't importing a foreign fitness fad; we were handing back something familiar — and rope skipping, done properly, pays you back out of all proportion to how simple it looks: real cardiovascular fitness, real coordination.

Why run it on WhatsApp? Same logic. Everyone already has it. No new app, no learning curve, nothing to figure out — you skip, you record, you send, on the tool already in your hand.

And underneath all of it, one more aim: to give these women something most programs forget — a community. Not a solo grind, but a place to belong, built around movement.

Meet the women

None of them had the time. All of them found it.

Nancy. At 56, she's one of our oldest skippers — soft-spoken, with a heart of gold and the quiet, relentless work ethic of an ant. She's only recently found her way back into fitness, and in May alone she logged 11,435 verified skips. That's more than 350 a day, every single day of the month. It puts her at number four on the board, and anyone watching her climb can see the top three aren't as far off as they look.

Dr Jane. She skips as Invicta — unconquered — which is the only handle that would fit her. She's a doctor: brilliant, with the heart of a lion, carrying her own family and everyone else's, because that's what the work asks. She's an athlete besides — she has represented Nyandarua in inter-county badminton, and runs a racket-stringing business on the side. Her morning begins like the others' — child up, ready, off to school — and then she's at the hospital: ward rounds on expectant mothers, then theatre, where on an average day three of her cases turn into emergency C-sections. Some nights she's pulled from bed at midnight for a delivery that won't wait. And in the gaps between the lives she saves, she found time to skip her way to number three and — for now — into the Order of Nzinga, the unconquerable. Invicta is right.

Carol. Carol came to win. One of the most fiercely competitive people I've ever met — and she runs Runda Academy, the school I trust with my own children. You'll catch her skipping in the schoolyard with her pupils cheering her on, a grown woman showing a generation of kids that movement is something you carry for life, not something you leave behind. She sits at number two with 32,710 skips, barely fifteen hundred behind the leader. And she's competitive enough to have broken the app: she kept hitting the video upload limit because she wanted to submit more, and forced us to raise the cap. Carol outgrew the tool, so we rebuilt the tool.

Ruth. Resilience with a timetable. Six days a week her job sits sixty kilometres away, round trip, on public transport. She has a small child. Her day starts at five: get herself ready, get her child ready for school, the long commute, a full day's work, the commute home, homework at the kitchen table — and then, in whatever the evening has left, she skips. After that she sits down to prepare content for her radio show. Ruth found her way through depression in the boxing ring, alongside her coach Lemid, who heads the Nakuru Amateur Boxing Club, and she carries that same fight into her Sunday show, Kaihuri ka Ugi. When Ruth uses the word resilience, she isn't borrowing it. And look at the board: right now, she's first. 34,160 skips — more than anyone else in the county — squeezed into the cracks of the most punishing day on this whole list. Whether she's still first by midnight is another question entirely.

How you climb

Here's what hooks people: you don't just pile up a number. You rise through an order of African queens, and every rung carries a name — and a woman — behind it.

You join the moment you register, in the Order of Moremi, for Moremi Ajasoro of Ile-Ife, who gave herself up to save her people. Your first verified skip makes you Idia, the Benin queen mother who led armies and counselled kings. After that you climb on your skips alone:

• Nandi — 500 skips. The Zulu queen who raised Shaka from nothing. Resilience.
• Makeda — 2,000. The Queen of Sheba, whose wisdom was legend. Knowledge.
• Amina — 5,000. The warrior queen of Zazzau, who pushed her borders for thirty-four years. Warrior spirit.
• Nzinga — 10,000. The Angolan queen who fought the Portuguese for thirty years and never bent. The unconquerable.
• Yaa — 25,000. Yaa Asantewaa of the Ashanti, who at seventy led her people into war against an empire. Fearless eldership.

Yaa is the highest anyone in Nyandarua has reached. On the 29th of May, two women crossed twenty-five thousand skips to claim it: Ruth and Carol, the rivals you've already met, locked at the top of the board and crowned together. Sit with whose name that order carries — a woman of seventy who refused to sit down — and then look at who earned it here: women over thirty, with children, jobs, and midnight shifts.

But here's the thing: I'm writing this on the last day of the month, and nothing is settled. Ruth leads Carol by barely fifteen hundred skips — a gap Carol could erase before midnight and seize the top crown for herself. And Invicta, our unconquerable doctor, sits just under the line, close enough to claim the Order of Yaa tonight and make it three. By the time you read this, it may already be decided. Or it may be happening right now, in the dark, one skip at a time.

And Yaa is not the top. Three more orders stand above her — higher, and still unclaimed. No one in the county has touched them. We won't tell you their names. But they're up there, waiting for the first woman bold enough to reach them.

This is only getting started

Here's the strange part: most of Nyandarua still doesn't know this is happening. In May alone, twelve women logged 115,595 verified skips — up from under six thousand the month before. Women are climbing a dynasty of queens, outgrowing the app with sheer effort, being crowned in the Order of Yaa — and it's stayed one of the county's best-kept secrets. That ends here.

Because this was never a one-off with a winner and a closing ceremony. It's a recurring monthly contest for women over thirty, built around a community that actually shows up for each other. Every month the board resets. Every month there's a fresh climb — from the Order of Moremi all the way up, with three orders at the summit that no one has reached yet, still waiting.

So it doesn't matter that you're only hearing about this now. Nancy began again at 56. Next month's leaderboard is empty. The only question left is whether your name is on it.

Ready? It's free. Save +254 140 823802 on WhatsApp, send the message JOIN 018 QUEENS, and follow the prompts. That's it — you're in.

A few years ago—before AI, before smartphones became extensions of ourselves—technology was still confusing to many of our parents. As millennials, we found ourselves in an unusual position: we were the translators of technology.

We taught them how to save contacts, send messages, and make calls on simple feature phones.

I remember one particular neighbor—Jairo (Jairus).

He would show up at our house late in the evening, often slightly drunk, holding what he believed was a “broken” phone. I must have been around six years old at the time, but to him, I was something special—a small technician with extraordinary skills.

He never came empty-handed. A packet of mandazis was his way of paying for the service.

He would explain the issue, frustrated: he couldn’t make calls. My mother, without hesitation, would point at me and say, “Fundi wako ako hapo”—your technician is right there.

I’d take the phone, and almost immediately, I’d notice the small airplane icon at the top of the screen. Flight mode.

Simple problem. Simple fix.

But where’s the fun in that?

Like any “professional,” I had a reputation to maintain.

Jairo would say, “Ona, dũgakorwo na ihenya”—take your time.

And I did.

I’d walk to a quiet corner, open my favorite game—Snake Xenzia—and start playing. I’d chase high scores while occasionally checking the battery level. When it dropped to one bar, that’s when the real “work” began.

I’d disable flight mode, remove the battery dramatically, then ask him to give it at least ten minutes.

When I finally handed it back, he would immediately make a call—usually to his wife—and proudly announce, “Gore ni mwaki!” (Gore is fire). That was my childhood nickname.

Payment confirmed. Mandazis enjoyed.

At the time, it didn’t feel like a big deal. It was just a small win, a harmless trick, maybe even a child’s creativity at play.

But looking back now, it means something different.

I wasn’t fixing phones. I was operating in a gap—between knowledge and ignorance, between exposure and unfamiliarity. Jairo wasn’t incapable; he simply didn’t know.

And today, I can’t help but see the parallel.

We are now standing in a similar moment in history—only this time, the gap is called Artificial Intelligence.

AI is no longer a futuristic concept. It’s here. It’s moving fast. And just like back then, there are two groups forming: those who understand it, and those who don’t.

Some people will adapt early. Others will hesitate. Many will ignore it—until it becomes unavoidable.

And then, just like Jairo with his phone, they’ll find themselves locked out of something that seems simple to others.

The difference?

This time, the stakes are much higher.

Careers, businesses, and entire industries are being reshaped. The next generation will grow up with AI the way we grew up with mobile phones—it will be natural to them.

But for us, this is a transition.

We have to unlearn, relearn, and push through the discomfort.

We are the bridge generation.

We carry the responsibility of understanding this shift—not just for ourselves, but for those who come after us.

Because one day, someone will look at AI the way Jairo looked at that phone—confused, frustrated, and locked out.

The question is: will you be the one holding the mandazis, or the one fixing the problem?

Learn AI.

If not for yourself, then for the next generation.

Three AIs, zero coding background, one real product, and a lot more confusion than people imagine.

On April 5, 2026, a WhatsApp bot for women aged 30+ went live for our movement series in Nyandarua County, Kenya.

Women can register on WhatsApp, submit rope-skipping and fitness videos, get reminders, appear on leaderboards, and compete for monthly recognition. The bot stores data, syncs with Google Sheets, exports CSVs, supports admin review, and handles real-world messiness like incomplete submissions, wrong formats, and people typing greetings in the way people actually greet in Kenya.

I built it.

I still cannot sit here and pretend I am now a software engineer. I am not. A few weeks before this project, I could not have explained the difference between TypeScript and JavaScript in any useful way. I have no computer science background. I did not go to bootcamp. I am a doctor, a fitness builder, and a person who spends a lot of time thinking about real people in real places, not software abstractions.

And yet, somehow, I built a working WhatsApp bot.

Not alone. With three AI assistants: ChatGPT, Gemini, and Claude.

This is the honest version of how that happened.

Why this had to be on WhatsApp

The project itself was not random.

I have been working on community fitness and preventive health through Global Fast Fit. One of the ideas I cared about most was creating a practical movement platform for women aged 30 and above in Nyandarua. Monthly participation. Measurable activity. Real structure. Something that could become both a useful program and, eventually, a valuable fitness dataset grounded in African reality.

The obvious instinct is to build an app.

That makes sense if your users live inside app stores, stable internet, email logins, and endless phone storage.

That is not the world I was designing for.

In Nyandarua, WhatsApp is the real operating system. If I wanted this thing to live in people’s hands, not just in my imagination, it had to happen there. So the product became a WhatsApp bot that could handle registration, consent, submissions, verification, reminders, and leaderboards inside a tool people already use.

That was the idea.

Then came the tiny issue that I did not know how to code.

Phase 1: ChatGPT helped me think like a builder before I could build

The first serious work I did with ChatGPT was not “write me a bot.”

It was more like: help me think.

What data should I collect? What should count as a valid submission? How many rope-skipping videos per day is fair? Should GFF Standard be unlimited attempts? Should we use GFF Standard or GFF Shuttle for the movement series? What should happen if a video does not show the full body? What does consent need to cover? What fields are useful now, and what fields might become valuable later?

This part mattered more than I realized at the time.

Because for a product like this, the rules are the skeleton. If the rules are weak, the code is just a faster way to create confusion.

ChatGPT was strongest here. It helped me turn a rough idea into a system with logic. It helped define user flows, contest rules, admin commands, data fields, verification logic, storage thinking, monthly cycles, and all the small decisions that make the difference between “nice idea” and “working program.”

It also introduced me to the technology stack. Node.js. TypeScript. Express. Supabase. S3. Twilio. These were not familiar words to me then. ChatGPT explained them patiently, repeatedly, and sometimes like a person teaching a village chief how a post office works.

This was also where I think I did something right as a non-developer: I kept pressing on edge cases.

Not theoretical edge cases. Human ones.

What if two people share one phone?
What if someone keeps submitting until they get lucky?
What if the video is trimmed?
What if the bot becomes too complicated for a first-time user?
What if the scoring logic encourages the wrong behavior?
What if our perfect technical design is a terrible fit for Nyandarua reality?

That phase was slow, but it was good slow.

The weakness was that ChatGPT could sometimes become too architectural when I needed something more brutal and practical. I needed, “create this file, paste this, run this command.” Sometimes I got a blueprint for the city when I needed directions to the nearest spanner.

Still, if ChatGPT had a role in this project, it was the architect. It helped me design the bones.

Phase 2: Gemini entered when the machine started coughing smoke

Once the project moved from rules into actual code and deployment, the mood changed.

Now I was dealing with broken routes, webhook confusion, logs that looked like ancient curses, and that classic software experience where you have many moving parts and none of them are moving in the correct direction.

That is where Gemini became useful.

Gemini was better when I had something concrete and ugly in front of me. Error logs. Failing routes. Broken request flows. State issues. It was less interested in the philosophy of the product and more interested in what exactly was on fire.

One of the most annoying bugs was painfully small. I was dealing with a webhook path mismatch around /webhooks/twilio versus /webhook/twilio. One missing letter. Hours gone. That is software. You build a cathedral and then discover the door is painted on the wall.

Gemini was also helpful in the stretch where Twilio started feeling like a relationship that had run its course. We had designed around Twilio early on, but eventually I shifted toward Meta’s WhatsApp Cloud API. That switch was not elegant, and it definitely was not cheap in terms of energy, but it ended up being the right move.

Another hard area was session state. The bot would lose the plot between messages. A user would answer question three and the bot would behave like it had never met her. We kept circling that until the idea of explicit user state became clearer to me: where is this person in the flow, what did they already submit, what is the next expected thing?

That sounds obvious when written neatly in a blog post. It did not feel obvious at midnight with a broken bot.

If ChatGPT helped me think in systems, Gemini helped me respect logs. That may actually be one of the biggest lessons of this whole project. People talk about prompting. Fine. Prompting matters. But log reading matters more. Once I learned to copy the exact error, not paraphrase it, not soften it, just paste it as-is, the AIs became dramatically more useful.

Phase 3: Claude came in when I needed a builder, not a philosopher

By the time I moved heavily into Claude, the project had history, scars, and fragments everywhere.

There were handover notes. There was partial code. There were missing pieces. There were earlier design decisions that no longer matched the infrastructure. There was at least one machine continuity problem. And there was that feeling every messy project gets, where some of it lives in the codebase, some of it lives in your head, and some of it lives nowhere at all.

Claude was strongest when the task became: inspect this thing, compare it to the spec, identify what is broken, and systematically work through the gaps.

That mode suited the project.

One of the biggest practical breakthroughs came around message handling and Meta integration. At one stage the bot could receive messages but could not send them properly. At another stage menu routing was wrong enough that users could get pushed back into the wrong path. There were also infrastructure headaches that had nothing to do with my intelligence or Claude’s intelligence and everything to do with the fact that Meta can be absurd.

I want to be very clear about that. Some parts were not “hard because I am a beginner.” They were hard because the platform itself was chaotic.

We had cases where a number looked present in the dashboard but behaved like it did not exist at the API layer. That is not a poetic failure of the human spirit. That is just platform nonsense.

Still, Claude was helpful in systematically untangling those parts. It was closer to a builder or auditor. Less “let’s brainstorm” and more “here are the violations, here is the sequence, now fix them.”

That mattered, because by then I needed momentum more than inspiration.

The hardest part was not coding

I know that sounds like a trick line, but it is true.

The hardest part was continuity.

That is the thing people do not understand when they fantasize about building with AI. They imagine you open a chat window, say “build my product,” and receive software like takeaway food.

That is not what happened.

What happened was that the project lived across multiple AIs, changing stack decisions, handover notes, a crash or at least a broken environment, and my own evolving understanding of what I was trying to build.

Every time I switched AI, I gained a fresh pair of eyes and lost context.

Every time I updated the architecture, I solved one problem and created documentation debt somewhere else.

Every time the machine failed or the environment changed, I was reminded that a system can be real in six places and still feel missing.

That was the hardest part. Not syntax. Coherence.

In hindsight, if I had done one thing better, it would have been this: maintain a brutal running document of what works, what is broken, what changed, what the current stack is, and what the next step is. Not a beautiful document. A war diary.

Because without that, building with multiple AIs starts to feel like running three relay races on three different tracks while carrying the same baton in your teeth.

What I think I did well, despite not being a developer

I was not bringing coding skill to the table. So the question is: what was I bringing?

A few things, I think.

First, I knew the users.

That sounds small, but it is not. The AI did not know Nyandarua women. The AI did not know what it means to build for people who are not living in product-demo land. The AI did not know what a confusing prompt feels like to a first-time user, or why a WhatsApp-native flow matters, or why asking for the wrong thing at the wrong moment makes the whole system feel alien.

I kept dragging the project back to real life. That was one of my main jobs.

Second, I was stubborn about logic.

I did not accept “it should work.” I kept asking, how exactly? What happens next? What if the video is wrong? What if the user goes silent? What if the same person tries again? What counts as fair? What breaks comparability? What data is worth collecting and what data is just decoration?

That helped.

Third, I got better at giving raw material instead of vague feelings.

When something broke, I learned to paste the exact error. The exact route. The exact response. The exact strange behavior. AI is much worse with “it’s not working” than with “here is the log, here is the request, here is what happened.”

I am convinced that this was one of the most important skills I developed.

What I did badly

Plenty.

I let version 1 stay liquid for too long. I was trying to think about the future, the dataset, the scale, the architecture, the monetization, the wider movement, and the immediate bug all at once. Vision is good. Too much simultaneous vision becomes fog.

I also switched contexts a lot. Sometimes that was necessary. Sometimes it was just expensive.

And like many non-developers, I occasionally wanted the answer to be conceptual when the truth was painfully local: wrong route, wrong token, wrong variable, wrong state, wrong order of operations.

Software is humbling in that way. It does not care that your idea is noble. It wants the right character in the right place.

The most memorable moment

There were a few.

The funniest category of moments was when the bug was microscopic and the suffering was enormous. A wrong path. A missing assumption. A state mismatch. That happened more than once.

The most frustrating moments were definitely around Meta. Fighting an API that behaves like it is gaslighting you is a special form of modern pain.

But the best moment was simple: the bot replying for real.

Not in theory. Not in a sandbox. Not in a document. Not in a handover plan. Actually replying.

That moment mattered because until then the project was a swarm of ideas, specs, logs, rewrites, and stubbornness. After that, it was a thing.

A real thing.

What I learned from using three AIs

ChatGPT was best for product design, architecture, structure, and forcing me to think clearly.

Gemini was best when I needed a mechanic, when the machine was already broken and I needed someone to stare at the ugly parts with me.

Claude was strongest when the task was systematic execution against a spec.

So yes, the AIs felt different. Not magical, just different. Different habits of thought. Different styles of usefulness.

But the larger lesson is that none of them could replace judgment.

They could generate code. They could explain concepts. They could debug. They could suggest architecture. What they could not do by themselves was care about the actual women this product was for, or decide what kind of experience made sense in this context, or hold the whole mission steady when the tooling got chaotic.

That part was still mine.

What I would tell another non-developer trying this

Start with the rules. Before you ask for code, define the real system.

Keep a running project diary. Every day. What works, what broke, what changed, what is next.

Copy exact errors. Exact logs. Exact routes. Exact outputs.

Freeze version 1 earlier than your ego wants to.

Do not confuse “big vision” with “current step.”

And most importantly, know your users better than the AI does. That is your leverage. The AI can generate the scaffolding. You have to make sure the building belongs in the neighborhood.

The honest conclusion

This project was not a clean triumph. It was not smooth, elegant, or cinematic.

It was fragmented. It was frustrating. It involved loops, rewrites, false starts, handovers, platform nonsense, and the constant feeling that the project might scatter if I stopped holding it together.

But it worked.

That matters to me.

Because on the other side of all the technical confusion is something very simple: women in Nyandarua can now use a familiar tool to join a structured fitness competition, submit performances, and be part of something larger than a spreadsheet or a speech.

And I got there without becoming a traditional developer first.

I got there by combining obsession with the problem, patience with the process, and three AIs that each helped in different ways.

So no, I cannot code in the normal sense.

But I have now shipped software.

That sentence still feels strange in my mouth. But there it is.

Global Fast Fit hiking club had me at the top of Mt Longonot.

Mount Longonot is a dormant volcano located in the Great Rift Valley near Lake Naivasha, rising to an elevation of 2,776 meters. It is a popular hiking destination featuring a 3.1 km steep ascent to the rim and a 7.2 km loop around the 1.8 km wide crater, offering scenic views, with a total hike time of 3-6 hours. Admittedly, it is not too long but it is very steep. As we went round the rim of the volcano, I was treated to stunning panaromic views including 3 different rainclouds, in three different areas, pouring rain and thunder. It felt God like. My mind started to drift. 

As you hike the trail, If you have a good pair of lungs and native Kalenjin legs carrying you, like I do, then you will use 50% of your energy going up the mountain and the other 50% noticing the memory of a massive volcanic eruption a long time ago. The long fissures on the ground, the lava flow remains, the deep crater left, the abundant ecosystem that has laid claim to these remains and attempt to hide the extent of damage done on this mountain by a violent series of eruptions. The eruption solidified its name, Oloonong'ot, as it created dramatic, rugged spurs and steep ridges that still define the mountain today- Mountain of Many Spurs. 

I started to think about how beginnings are rarely soft. Even life itself has painful, forceful beginnings. The universe does not favor gentle beginnings. Its tool of choice is the violent burst: rupture, pressure, tearing, collapse, eruption, burning, breaking. Beginnings are often misremembered as hopeful, peaceful, glowing things but when you actually look closely, many beginnings are brutal. 

A new life does not drift quietly into the world. Child birth comes through contractions, pressure, tearing, blood, pain, and immense force. For the chick to begin its life in the world, it must strike, crack, and break the very structure that once protected it. A seed does not grow by staying whole, the casing splits, the contained form breaks, and only then can roots and shoots emerge. New land can be born through volcanic explosion. The eruption looks destructive, yet it is also creative, laying down the material for future ground. Floods are feared for their force, but they also carry and spread rich sediment that later supports healthy regrowth. Forest fire consumes, but it also clears dead matter, opens cones, returns nutrients to the soil, and creates conditions for regrowth. Even something as fluid as water begins new landscapes through relentless force. Over time a river cuts, wears down, and reshapes stone itself. A butterfly emerging from a chrysalis is not neat. The butterfly must struggle out. The emergence is effortful because the struggle itself helps form what is needed for flight. Ease would actually interrupt development. Muscle develops when strain creates tiny tears in tissue, which the body then repairs into stronger form. An island may begin in hidden violence, with magma forcing its way upward beneath the sea. A beautiful island paradise begins in fire, pressure, and upheaval.

Inner life follows the same pattern. A person often does not awaken gently. Awareness begins when a lie fails, when denial can no longer hold, when an inner structure cracks. A new mind begins in the ruins of a false one. Often, emotional healing begins with emotions like anger, shame, confusion, pain or conflict.

I think the energy feels violent because beginnings are rarely clean. They are crowded with tension, uncertainty, and force. They push matter, body, time, or self out of one form and into another it is not just the start of something but rather, a rearrangement of something already in existance. This kind of change must have friction. It must cost something. One state has to break so that another can exist. 

Maybe we suffer partly because we keep expecting beginnings to feel gentle when the pattern all around us suggests they are often scorching. 

It was a calming thought. The hike may have been tough but I felt warm. I realized that the we often dread the intensity, the instability and the disruptive nature of life in general and use a lot of time and effort to keep things the way they are. But the universe is much bigger and stronger and will have its way every time. The universe concentrates force to break one form so that another can begin. So I do not need to mistake intensity for failure but, to stay steady enough to be remade as need be.

It’s midday but the sky is overcast like it's still morning. The city well known for its morning rain had a shower in the night and the signs are visible but this however doesn't deter the activities being done. Up at COBAP gym on the Lubya Hill in Kampala city, you can hear the brooms scraping against the new cement floor.  The gym apart from being larger than the neighbouring houses doesn’t look different on the outside than the general housing in the area. For anybody who has stayed in Kampala, Lubya hill on this side, seems an anomaly because the unspoken rule in Kampala is that the rich take the hills and poor take the valleys. Here however, the opposite seems true and it speaks loudly in the way the houses appear and how the gym used to look

A few months back, the gym was a bare murram floor, with a stack of car tyres, in the corner, a dingy but wide room with a slanting roof that would quickly look dark even with the gaping holes in its walls. The club boxers would come in at intermittent times to train under the watchful eye of the head coach popularly called Coach Lora, a simplified name for his long one, Lawrence Kalyango.

The gym has its roots from a community organisation that was helping AIDS victims. An organization called The Community Based AIDS Programme(COBAP)  had operated here and sports was part of their activities they did in the community and the boxing sessions were done in the gym. However, the organization died but the boxing continued and the name stuck. 

There is momentary stillness as the sweeping has been completed but there is a desire to make the place sparkle much more than than the sweeping could do. The renovations in the place are quite evident. A new cement floor, a sitting area, window frames already put in place, some brand new brickwork where there used to be empty gaps and a fresh coat of paint on the inner walls and  office. Overhead, are bigger bulbs for better lighting.  This renovation breathes  life and a new sparkle  into a gym that was known for producing  talent from a ghetto. 

There is a lull in the sweeping, water has to be purchased to clean the floor, the gym is about to get its first scrub, before that water was purchased to spray on the dust in the room so that it wouldn’t rise up during trainings, this time however, water will be used to make the floor sparkle. And this means some discussion before that can happen.

Just below the gym on one side, is a home with a tank where they sell water to fill a 20 litre jerrycan at five hundred shillings but on the upper side there is a new person selling three 20 litre jerrycans at one thousand shillings. The decision of who to purchase from is not that difficult to make. Every coin matters, especially in a place like this. And it doesn’t end there, when the water is poured, a cloth is placed at the entrance of the gym so as not to allow the water to flow outside. The water that collects there is poured back into the basin and reused to scrub the unscrubbed portions of the floor.. 

Every boxer who arrives just joins without being told. This is a team that knows the value of  team work. In the hive of activity one can see Batte Nuhu, just the other day, on Boxing day, he retook the Ugandan welterweight belt again. There is Kimera Moses who also won the Flyweight 51 kg category. These are big achievements and highly lauded but on this day here at COBAP, it seems like it's business as usual. The pomp and hype is nowhere to be seen, it's back to work like nothing has happened. The water has to be poured back into the basins, the brooms have to be pressed firmly into the new floor because you can bet people here want it sparkling, ready for its opening. After that, the punching bags will be lifted up for training to start.  Everybody participates regardless of status, belts or  kilograms. One thing one can see is the watchful eyes of Coach Lora who paces like the way he does when he is training them or by the ringside as they put into practice what he taught them. He issues instructions and what he wants done and hum of work continues with the occasional joke, laughter or conversation that pops up. 

A distance away is Sanyu Roberts, the Uganda GFF manager who has been behind the renovations. You could detect a smile on his face as he looks at what is happening. He seemingly seems lost in his thoughts. One could even imagine he could be seeing the smiles of those who sponsored this amazing work. From a dusty, old gym to something remarkable like this that will still be changing the lives of those who had few options in life. You can already see it in the new spring in their steps and the smiles as the work continues in earnest.

When all this is done, the exercises will begin, the punchbags will face the full wrath of those scrubbing hands, this time clothed in boxing gloves and coach Lora will issue commands under the watchful gaze of the bigger, unblinking bulbs overhead. The past victories have never made anybody forget here that there are more victories ahead.