TELISHA PALMER
My name is Telisha Palmer, and my research focuses on leveraging artificial intelligence to support the sustainable growth of aquaculture in response to rising global food demand. With a background in marine biology and data-driven environmental systems, I have been drawn to aquaculture as a critical frontier for sustainable food production. As wild fisheries decline and global populations grow, the need for efficient, scalable, and intelligent aquaculture practices becomes increasingly urgent. My work centers on applying AI to optimize farming operations while minimizing ecological impact.
In my current projects, I explore how computer vision, machine learning, and real-time sensor data can be integrated into aquaculture systems to monitor fish health, predict feeding behavior, detect water quality changes, and prevent disease outbreaks. I develop and test models that analyze image and sensor data to automate decision-making processes that traditionally rely on manual labor or periodic inspection. These technologies not only enhance productivity and yield, but also improve traceability, reduce waste, and support animal welfare—all essential components of responsible aquaculture.
Despite its potential, applying AI in aquaculture presents several challenges, including the lack of standardized data, limited digital infrastructure in rural farming areas, and the need for localized models that adapt to regional species and ecosystems. I address these gaps by collaborating with small- and medium-scale farmers, collecting diverse datasets, and focusing on interpretable AI systems that empower rather than replace human expertise. I believe that a sustainable AI-driven aquaculture model must be both technically robust and socially inclusive to ensure long-term adoption and impact.
Looking ahead, my vision is to help build intelligent aquaculture systems that are climate-resilient, economically viable, and accessible across the globe. I am currently exploring the integration of satellite imagery, blockchain for supply chain transparency, and reinforcement learning for closed-loop farm optimization. My mission is to bridge marine science with artificial intelligence to create sustainable seafood solutions that address both global nutrition and environmental resilience. Through interdisciplinary collaboration and field-level deployment, I aim to contribute to a smarter and more secure aquatic food future.




Domain Expertise Injection: Aquaculture involves complex fish biology, environmental ecology, and engineering controls. GPT-3.5 lacks deep pretraining on domain literature and field data, limiting recommendation accuracy for feeding, water management, and disease diagnostics.
Multi-Task Coordination: The system must optimize feeding, forecast health, alert for disease, and simulate scenarios. Fine-tuned GPT-4 jointly optimizes multiple loss functions, preventing interference across prompts.
Water Quality
Innovative solutions for monitoring and improving water quality.
Data Acquisition
Deploying sensors for real-time water quality monitoring.
Modeling Insights
Utilizing historical data for actionable farming recommendations.
Health Forecasting
Predicting fish health through advanced modeling techniques.
Scenario Simulation
Testing various environmental scenarios for optimal outcomes.