How Hugging Face Inference Endpoints, Jobs, and Buckets Power Search on Papers with Code
Papers with Code just rebuilt its search engine on Hugging Face infrastructure—here's how practitioners can use the same stack.
Why it matters
A case study in using Hugging Face's managed inference, job scheduling, and storage services to power semantic search at scale. Practitioners building similar search or retrieval systems can learn the architecture and trade-offs.
The key facts
10 to knowPapers with Code migrated search to Hugging Face Inference Endpoints + Jobs + Buckets
Use case: semantic search over academic papers and code
Demonstrates managed inference + async jobs + object storage integration
Architecture applicable to enterprise search and retrieval systems
Published as technical blog post with implementation details
Papers with Code integrated Hugging Face Inference Endpoints for search
Uses Hugging Face Jobs and Buckets for infrastructure
Focus is on search capability powering research discovery
Published Aug 2026 — infrastructure partnership/integration story
Hugging Face positioning as enabler of research tooling
Go to the source
Hugging Face Bloghuggingface.co