Research & Innovation
Sicelukwanda is a specialized Robotics and Machine Learning engineer focused on simulation (MuJoCo, PyBullet) and hardware integration. Their work demonstrates deep mathematical proficiency and architectural ingenuity in solving complex domain-specific problems, such as hardware obsolescence and procedural physics generation. While technically sophisticated, their projects exhibit a "research-first" mindset, prioritizing experimental functionality over production-grade maintainability or testing coverage.
The score reflects high-level research capability and domain expertise rather than production engineering polish. While testing and CI/CD are lacking, the complex simulation logic and hardware integration demonstrate advanced technical proficiency.
The profile presents a consistent and credible portfolio of research-grade software in a specific domain. While the code lacks production polish (tests, CI), the complexity of the logic and the depth of the metadata provide a clear picture of high technical competence.
Prioritizes exploration over polish
Codebase for the paper "Learning Dynamic Tasks on a Large-scale Soft Robot in a Handful of Trials"
An attempt at using the legacy intel movidius compute stick with modern pytorch
Imitation Learning for the MyCobot robot in PyTorch
A simple plotting utility for cartesian paths and standard deviation in pybullet.
A Jax imitation learning tutorial for the Robot Learning for Africa workshop 2023