AI agent development and hosting
Standard Agents
Standard Agents builds software for developers who want to put specialized AI agents to work in their applications and business workflows. Standard Agents Builder lets developers test agents locally and deploy them to managed hosting. Once an agent is running, developers can inspect individual conversation threads and trace a failed run to the tool call that caused it.
Builder implements the open Standard Agent Specification. An agent’s model, prompts, tools, and hooks are defined in TypeScript, so developers can package the complete agent and run it on a compatible runtime. One agent can call another as a tool. The specialist runs in its own thread with its own context, so a support agent can hand off a return to a specialist that has the tools for that job.
Standard Code applies the specialist-agent approach to software development. Its cloud coding sessions can keep running when a developer disconnects and resume from another connected machine. Standard Agents also develops dmux, an open-source terminal tool that runs coding agents in parallel. Each task gets a tmux pane and a separate Git worktree, so developers can review and merge each task’s changes independently.
2 talks
Newest firstLoading more talks and speakers…
1 speaker at AIE
Affiliations reflect their AIE event listings and recorded appearances, not necessarily current employment.

