Abstract: Context-based control inputs for a device are described herein. In an example, a device presents a first user interface (UI). The device determines, while the first UI is presented, a first user interaction corresponding to a first instance of a user input with the device. The device determines a first context associated with the first user interaction and a first control input based on the first user interaction and the first context. The device causes execution of a first action based on the first control input. The device determines a second user interaction with the device that corresponds to a second instance of the user input. The device determines a second context associated with the second user interaction and determines a second control input based on the second user interaction and the second context. The device causes execution of a second action based on the second control input.

Abstract: Techniques are disclosed for intelligently managing software development. In one example, a method for managing software development, includes receiving, by a computing system, a request to review source code written by a first developer, determining, by the computing system, a software skill set for the source code review, selecting, by the computing system, one or more selected source code reviewers from the pool of source code reviewers based on the software skill set and respective reputation scores for a pool of source code reviewers, assigning, by the computing system, one or more portions of the source code for code review to each of the selected source code reviewers, and determining, by the computing system, a consensus verification output on the code review based on review input from a majority of the selected source code reviewers.

Abstract: Techniques are disclosed for training a machine learning model to perform actions within an environment. In one example, an input device receives a declarative statement. A computation engine selects, based on the declarative statement, a template that includes a template action performable within the environment. The computation engine generates, based on the template, synthetic training episodes. The computation engine further generates experiential training episodes, each experiential training episode collected by a machine learning model from past actions performed by the machine learning model. Each synthetic training episode and experiential training episode comprises an action and a reward. A machine learning system trains, with the synthetic training episodes and the experiential training episodes, the machine learning model to perform the actions within the environment.

Abstract: Techniques are disclosed for intelligently managing software development. In one example, a method for managing software development, includes receiving, by a computing system, a request to review source code written by a first developer, determining, by the computing system, a software skill set for the source code review, selecting, by the computing system, one or more selected source code reviewers from the pool of source code reviewers based on the software skill set and respective reputation scores for a pool of source code reviewers, assigning, by the computing system, one or more portions of the source code for code review to each of the selected source code reviewers, and determining, by the computing system, a consensus verification output on the code review based on review input from a majority of the selected source code reviewers.

Abstract: Techniques are disclosed for training a machine learning model to perform actions within an environment. In one example, an input device receives a declarative statement. A computation engine selects, based on the declarative statement, a template that includes a template action performable within the environment. The computation engine generates, based on the template, synthetic training episodes. The computation engine further generates experiential training episodes, each experiential training episode collected by a machine learning model from past actions performed by the machine learning model. Each synthetic training episode and experiential training episode comprises an action and a reward. A machine learning system trains, with the synthetic training episodes and the experiential training episodes, the machine learning model to perform the actions within the environment.