How I automate my own job at Hugging Face using agents
This talk will showcase how I automated a large part of my own job at Hugging Face. This involves both open (GLM-5.1) and closed-source models (Claude, Gemini), the Claude Agents SDK, serverless infra like Modal and Hugging Face Jobs. I will also discuss how I use agentic coding tools like Cursor and Codex to implement AI agents which automate my job, and how everything is connected to the internal Slack of Hugging Face.
About the AI Architects: Show my Workflow Track
AI Architects: Show my Workflow sessions at AI Engineer World's Fair 2026 in San Francisco.
When
Tuesday, June 30, 2026
2:50 PM - 3:10 PM·20m
Where
Leadership 2 · Room 3020
Capacity: 550 attendees
Sign in to add this talk to your schedule.
Speaker
Niels Rogge
Hugging Face
@NielsRogge
Niels works as a Machine Learning Engineer at Hugging Face as part of the Community Science team. Together with @_akhaliq (also known as AK), which you might know from posting trending research papers, he ensures researchers improve the discoverability and visibility of their artifacts by making them available on the hub with proper links to the paper, write model and dataset cards and more.