Building at the Brink: How a 25 Year Old Is Using AI and Two Companies to Speed Coral Reef Rescue

Lead: A 25 year old founder who weathered academic failure, a serious car crash and dual infections has taken a data centric reef restoration report to the United Nations, while running a revenue generating vertical farming startup. Yusuf, CEO of CoralX Foundation and Jr Greens, is betting that computer vision, text aggregation and developer talent can accelerate coral conservation across 30 plus countries, and he has already shipped tools that identify coral species and health from dive footage.

What happened

Yusuf used a combination of AI models and traditional data pipelines to build a Reef Restoration Report that maps who is doing restoration work, how they operate, and what their outputs are. That report earned CoralX visibility at the UN Oceans Conference, opening doors to ministers, presidents and potential funders. CoralX is not just publishing status updates on reefs, it is building operational tools: computer vision that analyzes GoPro dive footage to label coral species, mark live versus dead colonies, and flag associated fish species, plus pipelines that summarize research literature for practitioners.

Parallel to CoralX, Yusuf runs Jr Greens, a vertical farming company that currently supplies revenue and bandwidth to support the nonprofit work. He describes a pragmatic split: the for profit funds his livelihood, the nonprofit focuses on community tools and open knowledge. Yusuf also emphasizes developer talent, noting contributors from major tech firms and AI backgrounds have helped build the tooling.

How the technology and model work

Computer vision: Divers collect video or images, the model processes frames and returns labels for coral type, health state, and fish presence. The goal is to collapse what would take days of manual annotation into automated reports.

Text aggregation: CoralX applies AI to parse research papers and extract actionable signals that are often buried in jargon, helping practitioners find novel findings faster.

Data pipelines: Outputs from vision and literature are aggregated into dashboards and regional reports that show capacity, methods, volunteer counts, species targeted, and funding profiles for restoration groups.

Yusuf frames the approach as developer led: experienced coders who understand robust system design and data hygiene are the ones producing the most reliable tools, while AI accelerates iteration and lowers the entry barrier for smaller teams.

Funding is mixed, with Jr Greens providing operational revenue for Yusuf, and CoralX seeking philanthropic donations to scale tools that are not yet commercializable. Yusuf says some of the CoralX tooling exceeds what can reasonably be turned into profit and will therefore rely on donors, institutional grants and partnerships.

Why this matters

Coral reefs are a critical resource for coastal fisheries, shoreline protection and biodiversity, yet they face rapid decline from warming, pollution and overfishing. Yusuf cites mainstream projections that reefs face large scale loss by mid century if current trends continue. The combination of fragmented practitioner data and limited monitoring bandwidth means policy makers and funders lack a clear picture of who is restoring reefs, how effective those efforts are, and where to allocate resources.

CoralX attempts to close that gap by making practitioner metrics visible, automating labor intensive monitoring tasks, and translating scientific literature into operational insights. If the tools work at scale, they could help governments and NGOs prioritize interventions, match partners on a local level, and reduce duplication across islands and coastal regions where multiple organizations currently work in silos.

There is also a broader signal here for climate tech and conservation: developer skills and applied AI matter for translating field work into policy relevant data. Yusuf points to engineers from large tech companies as part of his team, arguing those software practices can be repurposed for planetary scale problems.

What is uncertain or risky

Model accuracy and bias: Computer vision for underwater imagery faces domain shifts, from camera angles to turbidity to species variability. Models trained on limited datasets can misclassify coral health, which would mislead management decisions unless accompanied by rigorous ground truthing.

Adoption by practitioners: Many reef groups operate with limited budgets and training. Shipping a tool does not guarantee uptake. CoralX will need to prove ease of use, low cost of data collection, and clear benefits for local practitioners.

Funding durability: Yusuf expects some CoralX work will need philanthropic support. Long term sustainability depends on converting enough stakeholders into recurring funders or building hybrid revenue streams without compromising a public good mandate.

Integration with policy: Data alone does not change fishing rules, coastal development permits, or funding allocations. Turning insight into policy requires political will, legal instruments and local leadership, which are harder to engineer than software.

What to watch next

Product rollout across 30 plus coral countries: Yusuf says the next phase is deployment to many coral nations. Watch for announcements about pilot partners, public datasets, or APIs that let researchers and NGOs test the models.

Validation studies: Independent technical validation, ideally peer reviewed, that reports accuracy on diverse reef imagery will determine credibility among scientists and funders.

Partnerships with regional hubs: Uptake will depend on local partners who can translate model outputs into action. Look for collaborations with marine research institutions, coast guards, and regional conservation funds.

Funding moves: Will CoralX secure philanthropic multi year grants, or will it pursue a hybrid social enterprise model for certain tools? Funding choices will affect scale and openness.

Bottom line

Yusuf presents a compact thesis: mature developer talent plus applied AI can turbocharge conservation workflows that previously required expensive, time consuming manual labor. The Reef Restoration Report is a strategic move, because mapping the actors who actually restore reefs helps make the case for targeted investments. The approach is promising, but its impact will depend on model reliability, practitioner adoption, and stable funding. If CoralX can prove the technology in diverse real world conditions and convert insights into local action, it will offer a replicable roadmap for other environmental domains where monitoring and fragmented data are the bottleneck.

Source: Building At The Brink