Daniel Fried is an assistant professor at Carnegie Mellon University’s Language Technologies Institute. His group studies how people and language-model agents communicate and collaborate, including code, computer use and multi-agent interaction.[1]
Start with 11-768: AI Agents →
X · dan_fried Contributions
Fried and Graham Neubig teach CMU’s Fall 2026 AI Agents course. Students build a coding-agent harness, design evaluations and implement training procedures before pursuing a research project. The syllabus connects tool use and context management with planning, safety and interaction. It treats an agent as a system that must act over many steps, not simply produce one answer.[3]
Ideas
A recurring theme in Fried’s research is pragmatics: a person uses language in a situation to affect a partner’s understanding or actions. His work applies that perspective to code generation and collaborative agents. The question extends beyond whether an output is fluent to how it helps two participants complete a task together.[1][4]
Tools, writing & talks
Writing, research & talks
Start here
11-768: AI Agentscmu-agents.comA graduate course combining agent harnesses, evaluations, training and research projects.
Start with the overview and assignments. The course expects prior experience training neural language models; enrolled students receive course compute resources.
Co-instructor with Graham Neubig[3]A university course at Carnegie Mellon University focused on building AI agents end-to-end. Students gain hands-on experience constructing agentic harnesses, building evaluations, and training LLMs with reinforcement learning. The course balances theory and practice while covering modern frameworks and best practices.
Catalog workOriginal source