Hi, I'm Saifullah.

I've loved computers and video games for as long as I can remember, so programming came naturally. I started coding in 2012 with a C/C++ course at university, building a Reversi game project, and soon got into robotics, first aerial robotics, then humanoid robotics for my Bachelor's and Master's theses.

During my time working on robotics, I have worked on several areas; motion planning and control, robot kinematics and dynamics, computer vision, SLAM, and behavior planning. I also led my university's RoboCup team for Standard Platform League, coding autonomous humanoid robots to play football.

Afterwards, I joined the National Center of Artificial Intelligence (NCAI), Pakistan, leading projects in UAVs, robotics, and deep learning. During this time, I led projects integrating ROS, PX4, and reinforcement learning for UAVs, implemented GPU-accelerated obstacle avoidance algorithms, and deployed real-time computer vision systems with NVIDIA DeepStream. I also worked quite extensively with simulators such as Gazebo, VREP, AirSim, etc.

Now, as a PhD researcher, I focus on explainable AI, private ML, and robust deep learning for document analysis, with extensive use of PyTorch. During this time time I have explored a wide variety of deep learning areas, and my areas of expertise include multi-modal document analysis, Explainable AI, GenAI, Diffusion Models, Differential Privacy and Distributed Training, Active Learning, and more.

I have been coding for more than 10 years, however, there is still so much to learn and explore. I keep myself updated with the latest research and technologies, and I love contributing to open-source projects. When I'm not coding, I enjoy video games and anime!