I'm an AI Engineer and Research Fellow building intelligent systems that are interpretable, coordinated, and grounded in real-world applications.
- 🤖 Multi-agent LLMs and emergent coordination behaviors
- 🧠 Mechanistic interpretability and understanding model reasoning
- 🎯 Reinforcement learning with bounded rationality and cognitive constraints
- 🔍 Graph neural networks for structured reasoning and knowledge representation
- 🌍 Applied AI for climate adaptation, misinformation detection, and policy evaluation
- Agentic AI systems orchestrating LLMs with tools, search, and reasoning loops
- Multimodal AI combining vision-language models with graph-based evidence verification
- Foundation models fine-tuned for geospatial analysis, circular economy, and real-world deployment
- Production ML systems with RAG, vector databases, and MLOps pipelines
- Misinformation Detection: Agentic framework using LVLMs + GNNs achieving 93.05% accuracy
- Climate AI: Leading EU projects (CLIMATEAdaptEOSC, EcoPlast) with geospatial foundation models
- Supply Chain ML: Production forecasting systems deployed as FastAPI microservices
- MSc Artificial Intelligence @ Queen Mary University of London
- BSc Computer Science @ BRAC University
- Research Fellow @ Fatima Fellowship
- AI Engineer @ Technovative Solutions
Interested in research collaborations on multi-agent systems, interpretability, or responsible AI. Always excited to discuss PhD opportunities and impactful AI projects.