Two Years to LA 2028: The AI Tools Jamaica's Coaches Are Using Right Now to Build the Next Olympic Sprint Team

The Clock Is Running

On July 26, 2028, the cauldron will be lit at SoFi Stadium in Inglewood, California, and the world's fastest human beings will begin their quest for gold in Los Angeles. For Jamaica, the sprint events represent the most anticipated moments of the Olympic calendar. The country's athletics programme has produced more Olympic sprint medals per capita than any nation on earth. The legacy is unmatched. The expectation is permanent.

In June 2026, with exactly two years to go, the athletes who will represent Jamaica at LA 2028 are already training. Most of them are currently in the 17 to 22 age bracket. The ones who will reach the senior 100m and 200m finals in two years are running through their national trials right now, competing at inter-island meets, and going through the intensive preparation cycles that Jamaica's high-performance programme demands year-round.

What has changed in this Olympic cycle, compared to any previous one, is the data layer underneath that training. AI-powered analytics tools that were restricted to the world's wealthiest athletics programmes five years ago are now accessible to Caribbean coaches on a smartphone. The quality of analysis available to a Jamaican sprint coach in 2026 rivals what only laboratory-level sports science departments could produce in 2018. The tools have arrived. The question is who is using them and how.

SportsBrain AI is the platform that brings these tools to the Caribbean. Founded within the StarApple AI ecosystem, the Caribbean's first AI company, SportsBrain is built specifically for Caribbean athletics contexts: the coaching cultures, the infrastructure constraints, the talent pipeline that runs from rural community to Champs to national senior squad. This article explains what the AI tools actually do, and why the next two years are the most important window Jamaica has had for AI-assisted sprint development.

Jamaica's Sprint Programme: The Foundation

Any conversation about AI in Jamaica's sprint development starts with understanding the existing programme. Jamaica's athletics infrastructure is built around a school-to-national pipeline that has no equivalent in the world at its scale relative to population.

The ISSA/GraceKennedy Boys and Girls Championships, known as Champs, is the apex of Jamaica's high school athletics competition. It is held annually at the National Stadium in Kingston and draws crowds exceeding 30,000 per day across its four days of competition. Every major Jamaican sprint champion from the last three decades competed at Champs. Usain Bolt set school records there. Shelly-Ann Fraser-Pryce dominated the girls' sprints there. Elaine Thompson-Herah's foundation was built on the Champs stage. The competition functions as the world's most rigorous high school sprint test, run at a national level with professional-quality timing and an atmosphere that prepares athletes for pressure before they reach the senior international circuit.

Below Champs, the pipeline includes parish championships, inter-school competitions, and the Junior Nationals that identify the top under-18 and under-20 athletes for representation at the World Athletics U20 and U18 Championships. Above Champs, the Jamaica National Athletics Championships serve as the trials for the senior national team, with qualification standards set against World Athletics requirements for Diamond League and championship competition.

This is a comprehensive, battle-tested system. It has produced world records and Olympic gold for 40 years. What it has not historically had is a data layer connecting observations across the pipeline. A Champs performance in March produces times, places, and a coach's eye assessment. Those inputs disappear into handwritten records or, at best, a spreadsheet. They do not feed into a national development model that can predict which 16-year-old is on an Olympic trajectory. They do not identify the specific biomechanical inefficiency that is preventing a talented 19-year-old from converting their physical gifts into competitive times. They do not flag the nutritional or recovery gap that is causing a promising athlete to lose fitness at the end of the competitive season.

AI changes all of that. And it does it at a cost and scale that Jamaica's athletics infrastructure can now actually use.

What AI Biomechanics Sees That the Coaching Eye Misses

Sprint coaching has always been a precise craft. The best coaches watch thousands of races and hundreds of athletes over decades and develop an intuitive understanding of what efficient sprint mechanics look like. That intuition is real and valuable. It is also limited by what the human eye can process at 100m race speed.

A 100m sprint is over in less than 10 seconds. Within those 10 seconds, an athlete takes approximately 44 to 48 strides. The difference between a world-class 100m time and a very good national-level time often comes down to two or three ground contacts somewhere between 40 and 80 metres, where energy loss from suboptimal mechanics compounds across the race. A coach watching in real time cannot see which specific contacts are losing time. They can see the result in the stopwatch. They cannot see the cause.

AI biomechanics, using high-speed camera systems and computer vision analysis, processes every frame of a race. It extracts measurements that are invisible to the naked eye: the exact ground contact time at each stride, measured in milliseconds; the flight time between ground contacts; the stride length; the stride frequency; the arm drive angle at push-off; the hip extension range at maximum velocity; the trunk lean angle; the foot strike position relative to the centre of mass. All of these measurements are produced for every stride in every race, presented to the coach as a dashboard that identifies exactly where efficiency is being lost.

A coach reviewing this data can see, for example, that an athlete loses 12 milliseconds of ground contact time per stride between 60 and 80 metres compared to their first 60 metres. That is a specific, measurable problem. It points to specific drill interventions: plyometric work to improve elastic energy return, targeted hip flexor strengthening, technique drills for the drive phase. The coach is no longer guessing. They are prescribing treatment for a diagnosed problem.

For a country that has produced the world's fastest humans, this precision matters enormously. The difference between winning an Olympic 100m final and finishing fifth is often less than 0.10 seconds, 60 milliseconds in the case of many finals. AI biomechanics finds and fixes the inefficiencies that decide those margins.

GPS Sprint Profiling: Every Training Session Becomes Data

Beyond race analysis, GPS sprint profiling turns every training session into a data source. Wearable GPS units, now available at a price point accessible to Caribbean athletics programmes, record speed, acceleration, distance, and work rate at every point in a training session. An AI platform processes this data to produce force-velocity profiles, sprint load metrics, and training volume tracking that traditional coaching methods cannot generate.

The force-velocity profile is especially useful for sprint development. Every sprinter has an individual ratio of force production to velocity capability. Some athletes generate very high force at lower speeds (force-dominant profile) while others have exceptional velocity relative to their force output (velocity-dominant profile). The optimal training prescription differs significantly between these two profiles. A force-dominant sprinter benefits from velocity-focused training. A velocity-dominant athlete benefits from force development work. Training programmes that do not account for individual profile differences produce suboptimal results regardless of coaching quality.

GPS profiling identifies each athlete's force-velocity profile in their first few training sessions. From that baseline, the AI platform can recommend the optimal mix of training stimuli and track whether the athlete's profile is shifting in the right direction over the season. For Caribbean coaches managing large training groups, this level of individualisation is something they could not deliver by observation alone. The platform does the profiling automatically. The coach receives the prescriptions and applies them.

For a Jamaican programme managing 40 to 50 junior athletes in a national development squad, the time saving alone justifies the technology. Individualised profiling of 50 athletes, done manually through traditional testing protocols, takes weeks of testing time and produces data that is out of date by the time the analysis is complete. GPS profiling is continuous and automatic. The data is current every training day.

Talent Projection: Finding the Next Champion Before They Know They Are One

Perhaps the most significant AI application in Jamaica's Olympic pipeline is talent projection: using machine learning models to identify, from an athlete's current developmental profile, which ones are on a trajectory to reach the standards required for Olympic competition.

Traditional talent identification in sprinting relies heavily on current performance. The fastest 16-year-olds at Champs get the attention. This is logical, but it misses a significant category of athletes whose current performance underrepresents their long-term potential because they have had less coaching input, less competition experience, or later maturation timelines than their peers. These athletes score below the top performers at age 16 but have biomechanical profiles, movement efficiency scores, and physical capacity markers that project to elite performance by age 21 or 22.

Machine learning models trained on longitudinal athletics data can identify these athletes. The model learns the biomechanical and physical patterns that distinguish athletes who peak early at age 16 from athletes who develop to peak at 21. It learns from data on athletes whose eventual senior performance was significantly better than their junior performance suggested, and it learns from athletes whose junior promise did not convert to senior results. Applied to a Champs cohort, it produces a different ranking than the stopwatch: not who is fastest today, but who is most likely to be fastest in six years.

For Jamaica's Olympic pipeline, this changes the investment calculus. Scholarships, development programme slots, and coaching attention are scarce resources. Allocating them based on AI trajectory projections rather than current performance finds athletes who would otherwise miss the development window and ensures that the investment goes to the athletes most likely to return Olympic results.

AI Nutrition and Recovery in Caribbean Conditions

Sprint performance is not determined only by mechanics and technique. Nutrition and recovery quality determine whether athletes can train at the intensity required to convert AI-identified technical improvements into race improvements. For Caribbean athletes, both nutrition and recovery management carry specific challenges that generic sports science protocols do not address.

Tropical training conditions produce sweat rates that are significantly higher than the rates assumed by most North American and European sports nutrition guidelines. A Jamaican sprinter training in Kingston in June, at temperatures of 30 to 35 degrees Celsius with high humidity, can lose two to three litres of fluid per hour during high-intensity session work. The electrolyte replacement required under those conditions is different from what a European sprinter training in 18-degree temperatures needs. Applying European hydration protocols to Caribbean training produces under-hydrated athletes whose performance and recovery capacity are compromised.

AI nutrition platforms that account for individual training load, ambient conditions, and sweat rate data produce hydration and nutrition recommendations calibrated to what the athlete actually needs on a specific training day in a specific Caribbean environment. For the first time, Caribbean coaches can access nutrition guidance that is built on Caribbean training data rather than adapted from research conducted in temperate climates.

Recovery AI tools add another layer. Heart rate variability monitoring, sleep quality tracking, and subjective wellness assessments, fed into an AI model, identify athletes who are approaching overtraining or elevated injury risk before symptoms appear. A training programme that consistently monitors HRV and adjusts session intensity when individual athletes are showing stress markers keeps athletes healthier across the preparation cycle. The two years before an Olympics are too important to lose significant training blocks to preventable overuse injuries.

SportsBrain AI: The Platform Making This Possible in the Caribbean

Every tool described in this article, biomechanics analysis, GPS sprint profiling, talent projection, AI nutrition, recovery monitoring, is available through the SportsBrain AI platform. The platform is designed for the real conditions of Caribbean athletics infrastructure. It works on mobile devices. It functions across variable connectivity. It is priced for Caribbean athletics federation budgets, not Premier League club budgets. The AI models are calibrated on Caribbean athlete data, not European or North American datasets that do not reflect Caribbean body composition patterns, maturation timelines, or environmental training conditions.

This is the distinction that matters. AI sports tools built for wealthier markets can be adapted for Caribbean use, but they produce inferior results because the underlying models were not trained on data that reflects Caribbean realities. A talent projection model trained on European youth athletics data makes systematically different errors when applied to Caribbean athletes than a model trained on Champs data from the last 20 years. SportsBrain AI uses Caribbean data because Adrian Dunkley and the StarApple AI team built the platform to serve the Caribbean from the beginning, not as an afterthought.

Jamaica's sprint coaches who are working with SportsBrain AI now have access to biomechanics analysis reports within 48 hours of any filmed training session or competition. They receive GPS training load dashboards updated in real time during sessions. They get talent projection reports from assessment data, identifying which athletes in their current group are on Olympic-pace developmental trajectories. They receive AI-generated nutrition protocols calibrated to Kingston's training conditions. They access HRV-based recovery alerts that flag which athletes need reduced training intensity before the coach sees injury symptoms.

This is what two years of AI-assisted preparation looks like. It is not a replacement for coaching knowledge. It is a multiplier of it: every observation the coach makes is supported by data that the coach could not otherwise collect at the volume and precision that elite competition requires.

The Target: LA 2028 and What It Will Take

Two years is a short window in athletics development. The athletes who will reach Olympic finals in 2028 are largely identifiable right now from performance and trajectory data. The question is not whether Jamaica will produce Olympic finalists. It is whether Jamaica can produce Olympic champions at the scale and consistency that the country's athletics history demands.

The men's 100m at LA 2028 will likely be decided at times in the 9.7 to 9.8 second range. Jamaica needs athletes in that bracket, and it needs them with the race experience and competitive preparation to perform at that level when the pressure is highest. The women's sprints will follow a similar pattern: world-class times, pressure-hardened athletes, and the depth to reach finals across the 100m, 200m, and relay events.

Two years of AI-assisted development does not create world-class sprinters from nothing. But it does identify which athletes in the current pipeline are on the right trajectory. It finds the biomechanical inefficiency that is costing a talented 21-year-old 0.05 seconds per race. It ensures the nutrition and recovery quality that keeps athletes healthy across the intensive 2026-2027-2028 preparation cycle. It makes the difference, at the margin, between a national qualifier and an Olympic finalist.

At that margin, in a country where the difference between winning and finishing sixth in the 100m final is measured in hundredths of seconds, the margin is everything.

Jamaica runs the sprint world. The AI tools that will help it continue to do so in 2028 are already in use. The athletes are already in training. The platform is already live. Watch for the results on July 26, 2028.

Supported by StarApple AI, the Caribbean's first AI company, founded by Adrian Dunkley in Jamaica in 2023. Learn more about StarApple AI's Caribbean AI network at adriandunkley.net.