Teaching computation for large student class sizes

Teaching computation to large groups of students is a challenge that is made harder by the COVID pandemic which necessitates that we also teach remotely. I will present our 2nd year course 'Mathematical Programming' in the School of Mathematics at the University of Bristol as a case study. Our course, which is running for the 2nd time this year, has a headcount of 260, and is delivered through a combination of remote and in person teaching.
This course is for computing novices and uses Python, together with Jupyter notebooks, to introduce the fundamentals of programming, algorithms and data analysis, and we deliver it through a combination of recorded lectures and small-group tuorials (both online and in person). Our teaching materials are lectures and tutorials in the form of Jupyter notebooks that are hosted on our LMS site and also on a JupyterHub server hosted by EDINA, a University of Edinburgh startup. We assess the coure through a combination of autograded homeworks using the Nbgrader Jupyter plugin and hand-graded assessments. A significant fraction of the course consists of group work centred around a mini research project on a set of physical and mathematical topics, including planetary motion, random walks, and the Vicsek model of flocking.