She Watched Coaches Miss What Their Own Eyes Could Not Catch. So She Built an AI That Never Blinks.

She Watched Coaches Strain to See What They Were Missing. Then She Built the Tool That Sees It For Them.

A good volleyball coach watches a player's arm swing a hundred times before they ever say a word about it. They are looking for the small break in form, the shoulder that drops half a second too early, the wrist angle that costs a player power on every single spike. The best coaches catch these details. Most coaches, running drills for twenty players at once, catch maybe half of them.

That is not a failure of coaching. It is a failure of bandwidth.

A coach has one pair of eyes and a court full of athletes all moving at once. They are tracking footwork, calling corrections, managing timing, and reading the energy of the group simultaneously. The fact that a coach can observe as much as they do under those conditions is already remarkable. But remarkable is not the same as complete. And in sports development, the corrections that do not get made are the ones that compound quietly over time, becoming habits, becoming ceilings, becoming the reason a talented athlete never quite reaches what they were capable of.

Faith Okereke noticed this gap long before she ever wrote a line of code. She built Coach Vision to close it.

What Coach Vision Is

Coach Vision is an AI powered coaching tool that uses MediaPipe pose estimation to track an athlete's body movement in real time. Point a camera at a player during a drill, and the system maps their joints, tracks their motion, and flags the technical breakdowns a coach might miss in a live session. It turns a single camera into a second set of eyes that never gets tired, never gets distracted, and never looks away.

The system is not trying to replace what a coach sees. It is designed to extend it. When a coach is watching a player spike, they are simultaneously reading the player's approach angle, their jump timing, their arm extension, and a dozen other variables all at once. Coach Vision takes some of that load off the coach's attention and handles it computationally, surfacing the technical details that matter so the coach can give higher-quality, more targeted feedback in the time they have.

The result is a coaching environment where more gets seen, more gets corrected, and athletes get the individualized attention that their development actually requires rather than the fractional attention that a typical training session allows.

Faith built the entire project solo. That means she was designer, developer, and product tester all at once, working through the considerable technical challenge of getting pose estimation accurate enough to be useful for real athletes on a real court, not just a demo in a quiet room with ideal lighting and controlled conditions.

The Technical Foundation

Coach Vision is built on MediaPipe, Google's open-source framework for real-time body and hand landmark detection. MediaPipe identifies the position of key joints and body landmarks across video frames with high precision, and it does so without requiring specialist hardware. A standard camera or smartphone is sufficient. No motion capture suits, no depth sensors, no expensive equipment that a Nigerian sports programme would struggle to budget for.

This matters enormously. Sports technology in many parts of the world exists behind a paywall that makes it accessible only to professional programmes with professional budgets. High-performance video analysis tools, force plate measurements, GPS tracking vests, these are the tools of elite sport. Community coaches working with promising young athletes in Nigerian schools and clubs have almost none of them.

Faith chose MediaPipe specifically because it brings serious analytical capability to standard, affordable hardware. The same camera a coach might already be using to record training sessions becomes, with Coach Vision, a real-time analysis tool. That is not just a technical choice. It is a philosophical one about who this product is actually for and who deserves access to the kind of support it provides.

The pose estimation layer maps the athlete's body across frames, identifying joint positions and tracking how they change throughout a movement. For a volleyball spike, that means the system can trace the arc of the arm, the angle of the elbow at the point of contact, the rotation of the shoulder, the alignment of the hips, the height of the jump, and the landing posture, all from a single camera angle, all in real time.

What the system does with that data is where coaching intelligence comes in. Coach Vision does not simply produce a map of body positions. It flags deviations from correct technical form, the moments where what the athlete's body is doing diverges from what it should be doing to produce the outcome they are trying to achieve. That flagging is what transforms raw data into actionable coaching feedback.

The Observation That Became a Product

The idea came from a simple and honest observation. Athletic development in Nigeria often depends entirely on how many hours a coach can spend with each individual player. Talented athletes get overlooked not because they lack ability, but because there are not enough eyes and not enough hours in a training session to catch every correction they need.

This is a problem that scales with the size of the group. A coach working one on one with a single athlete can observe everything. They can watch every rep, give feedback on every movement, and adjust their guidance in real time based on what they see. But no sports programme operates on a one-to-one ratio. Coaches work with groups. Sometimes small groups, sometimes very large ones. And as the group grows, the quality of individual attention necessarily decreases, not because the coach is less skilled or less committed, but because their attention is a finite resource being divided across more people.

Coach Vision does not resolve that by adding more coaches. It resolves it by making the attention each coach has go further. When the system is tracking movement and flagging technical issues in real time, the coach is freed from having to catch everything through observation alone. They can focus on the things that require human judgment: the motivational state of the player, the situational context of a mistake, the nuanced communication that turns correction into development. The technical tracking, Coach Vision handles.

That division of labour between what the technology does well and what human coaching does well is not accidental. It reflects the understanding of someone who spent time in athletic environments before they became a developer. Faith did not approach this as a technology problem. She approached it as a coaching problem that technology could help solve.

What It Means to Build Something You Understand From the Inside

There is a meaningful difference between building a product for a market you have researched and building a product for a world you have lived in. The first produces something technically correct. The second produces something that actually fits the shape of the problem.

Faith spent time in the world of athletic coaching before she ever considered building Coach Vision. She understood the rhythm of a training session, the pressure a coach feels when they are trying to give every athlete what they need in the limited time available, and the particular frustration of knowing that a player has a technical problem you cannot quite pin down in the middle of a live drill. She knew what a coach was straining to see. She had watched coaches strain to see it.

That knowledge shaped every decision she made about what Coach Vision should do and how it should do it. The system is not designed to produce impressive visualisations or generate analytical reports that take an hour to interpret. It is designed to surface the specific information a coach needs in the moment they need it, in a format they can act on immediately. The design reflects an understanding of how coaching actually works rather than how a technologist imagines it works.

That is one of the things that makes Coach Vision genuinely interesting as a product. It is not the most technically complex project to come out of Renaissance Innovation Labs. But it is one of the most precisely aimed. The problem it solves is specific. The solution is specific. And the specificity comes from a builder who knew the territory before she drew the map.

The Broader Context of Sports Tech in Nigeria

Nigeria has produced remarkable athletic talent across virtually every sport it has seriously engaged with. The potential is not in question. What has always been less consistent is the quality and consistency of the technical development infrastructure that turns raw talent into refined performance.

Coaching quality varies enormously across regions, clubs, and school programmes. Equipment is often limited. Video analysis is rare outside of professional environments. The feedback loops that help an athlete understand and correct their technical errors are frequently slow, anecdotal, and incomplete.

This is the environment Coach Vision is designed to work in. Not in a funded professional academy with specialist staff and a video analysis budget, but in a community training environment where a coach with a smartphone and a group of motivated athletes could have access to tools that, until now, existed only for programmes far better resourced than theirs.

The potential reach of that is significant. Every sport has technical fundamentals. Every athlete develops faster when the feedback they receive is precise and timely. And every coach is constrained by the same finite attention that Faith identified as the core problem. Coach Vision addresses that constraint in a way that is accessible, practical, and built specifically for the context it is designed to serve.

Built Here, For Here

Coach Vision was built inside Renaissance Innovation Labs, in a community that does not let builders stop at interesting ideas. The lab pushes the question of what the idea actually becomes when it is real, when it is deployed, when real athletes and real coaches are using it in real training sessions.

Faith answered that question by building the whole thing herself, making every design decision, solving every technical problem, and testing the product against the standard of whether it would actually help a coach do their job better. That is a high standard. It is the only standard that matters.

Renaissance Innovation Labs exists to support exactly this kind of build. Not software for software's sake, but tools built by people who understood a problem from the inside before they ever tried to solve it. Faith spent time in that world. She knew exactly what a coach was straining to see. Then she built the thing that helps them see it.

If you work in athletic coaching, sports education, or youth development in Nigeria, Coach Vision is worth your attention. Not because it is finished or because it solves every problem, but because it is proof that serious, purposeful sports technology can be built here, by a builder from here, for the athletes and coaches who need it most.