Andrew Ng ships Open CoWorker desktop AI agent (github.com)

Andrew Ng has launched OpenCoworker, a cutting-edge desktop AI agent designed to enhance productivity by performing tasks and conducting research directly on users' computers. Built on the lightweight aisuite library, OpenCoworker can read files (with user permission), manage messages across platforms like Slack and email, and produce deliverables, including PDFs and spreadsheets. It also features scheduled automations, such as daily news summaries. Users need to supply their own API keys from providers like OpenAI, Anthropic, or Google, or run it locally with Ollama to ensure data privacy, with all data remaining on the user's machine. This release is significant for the AI/ML community as it democratizes access to powerful AI capabilities by providing a user-friendly interface that supports multiple language models through a unified API. Developers can easily create tailored agents by utilizing the comprehensive Agents API, which allows models to execute Python functions as tools, facilitating complex interactions and automations. The design prioritizes modularity, enabling users to integrate new provider adapters effortlessly. OpenCoworker and aisuite represent a substantial advancement in desktop AI agents, potentially streamlining workflows across various sectors by marrying advanced AI functionalities with everyday computing tasks.