AI PhD Student | Deep Learning Researcher | Polyglot Developer

I'm a passionate AI researcher working at the intersection of deep learning, computer vision, and natural language processing. My research leverages PyTorch and Hugging Face Transformers to tackle real-world challenges in satellite imagery and spatio-temporal forecasting.

When I'm not training models, I build high-performance systems using Rust, Go, and Kotlin, always aiming for efficiency, safety, and scalability.

• AI/ML: PyTorch, Transformers
• Languages: Python, Rust, Go, Kotlin
• Tools: Git, Linux, Docker, Jupyter

• Deep Learning & Representation Learning
• Computer Vision (Satellite Imagery)
• Natural Language Processing & Causal Modeling
• Efficient Neural Architectures

ICCV 2025
👉 Paper Link We propose a hybrid architecture that leverages the global context of Transformers and the local feature extraction power of CNNs to achieve state-of-the-art efficiency and accuracy in detecting objects in large-scale satellite images.

TMLR 2025
👉 Paper Link We reformulate vessel trajectory prediction as a language modeling task using Uber’s H3 geospatial indexing system and adapt Causal Language Models to capture complex navigational patterns with strong generalization.

Open to research collaboration, open-source contributions, or technical discussions!

"The best way to predict the future is to invent it." – Alan Kay