Chip Huyen
Author of Designing Machine Learning Systems and AI Engineering.
2 evidence links
← Discover
Retrieval / RAG
Hire for retrieval that reports a number
Indexing, chunking, reranking, and grounding answers in a corpus.
8Published profiles in this specialty
100%Of those rows attach evidence
8Capabilities in the taxonomy
Why hire through this index
Retrieval is where most “chat with your docs” projects stall. This page lists published profiles that name RAG as a practiced capability and attach evidence.
Available retrieval / rag
8 published profiles listing this capability.
Erik Bernhardsson
Modal founder; Annoy and Luigi — vector search and pipeline plumbing.
2 evidence links
Eugene Yan
Applied scientist; the patterns catalogue for LLM-based systems.
2 evidence links
Jason Liu
Instructor — typed, validated structured output from language models.
2 evidence links
Jerry Liu
LlamaIndex co-founder; retrieval and document workflows over private data.
2 evidence links
Omar Khattab
ColBERT and DSPy — late-interaction retrieval and compiled LLM pipelines.
2 evidence links
Shreya Shankar
DocETL, and research on operationalising LLM pipelines over messy data.
2 evidence links
Ada ExampleDemo
Retrieval systems with an eval harness, not a demo notebook.
Remote · Selectively available · 2 evidence links
Frequently asked
Do these profiles cover evaluation pipelines?
Look for the evaluation capability next to retrieval, and open the evidence links. Senior RAG work in this index is expected to show a golden set or an eval toolkit, not only a chat UI.
Can someone migrate a failing vector store?
That is a common engagement. Filter Discover for retrieval, then read the evidence — migrations only show up if the engineer attached proof of the work.
What about hybrid search and reranking?
Production retrieval on this directory is described as hybrid (lexical + dense) with a rerank step. Pure vector search is rarely the whole answer at scale.