Heart Disease Prediction
A classification model that estimates cardiovascular risk from clinical and lifestyle indicators, with feature analysis and model comparison.
I develop intelligent systems and scalable software, with a focus on Machine Learning, Deep Learning, Generative AI, and Intelligent Agents.

I'm an AI and software engineer, and I approach my work with curiosity and care. When an idea takes shape in my mind, I follow it all the way through to something built and genuinely used.
My hands-on experience spans machine learning, deep learning, computer vision and software development. I start with cleaning and preparing the data, then design, train and evaluate the model, and stay with it through deployment and integration into real-world applications. What I enjoy most is working with real data in all its complexity and messiness, and integrating AI APIs into tools that are genuinely useful.
I have a research-driven mindset. I've contributed to academic research, peer-reviewed papers for scientific conferences, and taught and served as a teaching assistant, and explaining concepts to others sharpens my own understanding. I've also built AI and software solutions in industrial settings, where a project can't just work well, it has to be reliable and maintainable. My goal is to turn AI into a practical solution for real-world problems.
Curious by default, precise by habit.
Engineering, research and teaching roles across AI, robotics and software.
Developed AI/ML solutions across financial market analysis, NLP, workflow automation and generative AI applications.
Software development, system design and implementation of organizational solutions.
Deliver hands-on training in AI tools, automation and prompt engineering.
Teaching assistant for the graduate-level Deep Learning course (MSc & PhD), supporting coursework and practical sessions.
Course repositoryPeer review of submitted papers for two academic conferences.
Programmed a quadcopter with Arduino — signal processing, motor and sensor control, PID tuning. Collected and annotated data and trained YOLOv8 for real-time object detection.
From real-time computer vision to predictive models trained on financial and biological data.
A classification model that estimates cardiovascular risk from clinical and lifestyle indicators, with feature analysis and model comparison.
Real-time hand tracking mapped to cursor control.
A CNN trained on facial imagery with augmentation.
Exploring transaction patterns to flag anomalies and summarise spending behaviour.
Sequence features turned into a classifier over genomic data.
Text classification on headline data with embeddings.
Predicting first-stage landing outcomes from launch data.
Available for collaboration or discussions about projects and technical questions.