As advances are made in robotic hardware, the complexity of tasks they are capable of performing also increases. One goal of modern robotics is to introduce robotic platforms that require very little augmentation of their environments to be effective and robust. Therefore the challenge for a roboticist is to develop algorithms and control strategies that leverage the knowledge of the task while retaining the ability to be adaptive, adjusting to perturbations in the environment and task assumptions. This work considers approaches to these challenges in the context of a wet-lab robotic assistant. The tasks considered are cooperative transport with limited communication between team members, and robot-assisted rapid experiment preparation requiring pouring reagents from open containers useful for research and development scientists. For cooperative transport, robots must be able to plan collision-free trajectories and agree on a final destination to minimize internal forces on the carried load. Robot teammates are considered, where robots must reach consensus to minimize internal forces. The case of a human leader, and robot follower is then considered, where robots must use non-verbal information to estimate the human leader's intended pose for the carried load. For experiment preparation, the robot must pour precisely from open containers with known fluid in a single attempt. Two scenarios examined are when the geometries of the pouring and receiving containers and behaviors are known, and when the pourer must be approximated. An analytical solution is presented for a given geometry in the first instance. In the second instance, a combination of online system identification and leveraging of model priors is used to achieve the precision-pour in a single attempt with considerations for long-term robot deployment. The main contributions of this work are considerations and implementations for making robots capable of performing complex tasks with an emphasis on combining model-based and data-driven approaches for best performance.