Skills · Build systems around models
Retrieval and RAG
Taught here
Embeddings, search and finding which stage of a pipeline failed.
Mentioned in 17 of 338 open research roles (5%) and 169 of 1,837 technical roles (9%), as of October 2, 2026.
Everything here that teaches it
7 pieces. Pieces built around this skill come first in each list; the rest use it along the way.
Book chapters
-
Chapter 5: Memory and retrieval: what the model sees, and a memory that forgets
Treat memory as choosing what the model sees: derive BM25 and the ranking metrics, and measure retrieval against sending everything on a real model. Then build durable, correctable preferences.
-
Chapter 6: Embeddings and vector search: when words are not enough
Build word vectors from scratch, then measure a real embedding model against BM25 on paraphrased questions: recall@3 rises from 0.2 to 0.733. Store each vector with its model and search a graph index.
Articles
-
Find the Failing RAG Stage Before You Change the Prompt
Measured on SciFact with small open models: attribute each miss to retrieval, reranking, prompt assembly or generation, then change only the stage that failed. A targeted prompt fix took 107 correct to 151.
-
Key a Cached Answer to Its Reader's Current Access, Not Just the Query
An answer cache keyed on the query serves a revoked reader their old answer. Add an authorization revision to the key, recheck before release, test the race, and count what per-reader keys cost in tokens.
-
Filter Retrieval by Permission Before Anything Reaches the Prompt
Permission-aware retrieval keeps unauthorized passages out of the ranked candidates, not just out of the answer. Test it on the prompt bytes, and measure what post-filtering costs in recall@k.
-
Measure Stale Agent Memory as Recall@k, Then Retire It by Source Revision
Similarity search ranks by meaning, not currency, so a note that used to be true can outrank the one that is. Measure it as stale@1 and recall@k of the current note, then retire by source revision.
Also in Build systems around models: Agents and tool use, Working with coding agents, Safety, red-teaming and misuse.
Who asks for it
- Research scientists and engineers 17 of 338
- Fellowships, residencies and internships 3 of 44
- Applied AI and forward-deployed 79 of 323
- ML systems and infrastructure 1 of 109
- Machine learning engineers 11 of 146
- Software engineering at labs 58 of 877
Open roles that mention it
169 open postings mention it. Here are the newest 60 research, fellowship, applied and ML-systems roles; the rest are on the role pages linked above. Each title opens the lab's own posting.
- Senior Applied AI Engineer ↗
CodeRabbit · Applied AI and forward-deployed · No degree stated · $200k to $300k base
- Partner Deployed Engineer ↗
Cognition · Applied AI and forward-deployed · No degree stated
- Partner Deployed Engineer - APAC ↗
Cognition · Applied AI and forward-deployed · No degree stated
- Partner Deployed Engineer - APAC ↗
Cognition · Applied AI and forward-deployed · No degree stated
- Partner Deployed Engineer - Europe ↗
Cognition · Applied AI and forward-deployed · No degree stated
- Research Engineer ↗
Factory · Research scientists and engineers · Bachelor's degree
- Research Engineer ↗
Greptile · Research scientists and engineers · PhD preferred · $180k to $280k base
- Senior AI Engineer (Core Engine) ↗
JetBrains (Junie, Mellum) · Applied AI and forward-deployed · No degree stated
- Machine Learning Scientist ↗
Relace · Research scientists and engineers · Master's, PhD or equivalent
- Forward Deployed Engineer ↗
Replit · Applied AI and forward-deployed · No degree stated · $180k to $300k base
- Applied AI Engineer ↗
Anthropic · Applied AI and forward-deployed · No degree stated
- Engineering Manager, Research Data Platform ↗
Anthropic · ML systems and infrastructure · No degree stated · $405k to $850k base
- Research Engineer, Knowledge Team ↗
Anthropic · Research scientists and engineers · No degree stated · $350k to $850k base
- Forward Deployed Engineer (India) ↗
Cartesia · Applied AI and forward-deployed · No degree stated
- Forward Deployed Engineer, Agentic Platform ↗
Cohere · Applied AI and forward-deployed · No degree stated
- Forward Deployed Engineer, Agentic Platform ↗
Cohere · Applied AI and forward-deployed · No degree stated
- Forward Deployed Engineer, Agentic Platform (Europe) ↗
Cohere · Applied AI and forward-deployed · No degree stated
- Forward Deployed Engineer, Agentic Platform (Singapore) ↗
Cohere · Applied AI and forward-deployed · No degree stated
- Forward Deployed Engineer, Agentic Platform (UK Public Sector) ↗
Cohere · Applied AI and forward-deployed · No degree stated
- Forward Deployed Engineer, Agentic Platform (West Coast) ↗
Cohere · Applied AI and forward-deployed · No degree stated
- Forward Deployed Engineer, Sovereign AI ↗
Cohere · Applied AI and forward-deployed · No degree stated
- Research Internship (Winter 2027) ↗
Cohere · Fellowships, residencies and internships · PhD required
- Senior Director, Solutions Architecture — Americas ↗
Cohere · Applied AI and forward-deployed · No degree stated · $270k to $540k base
- Senior Director, Solutions Architecture, APAC ↗
Cohere · Applied AI and forward-deployed · No degree stated
- Solutions Architect ↗
Cohere · Applied AI and forward-deployed · No degree stated
- Research Developer, Frontier AI Incubation, DeepMind ↗
Google DeepMind · Research scientists and engineers · Bachelor's or equivalent
- Research Scientist/Engineer, Frontier Reasoning, DeepMind ↗
Google DeepMind · Research scientists and engineers · PhD preferred
- Research Leader / Senior Scientist (Computational Drug Design), Cambridge, MA ↗
Isomorphic Labs · Research scientists and engineers · PhD or equivalent experience
- AI Research Engineer ↗
Meta (Superintelligence Labs and FAIR) · Research scientists and engineers · PhD or equivalent experience
- AI Research Scientist — Agentic AI for Materials Discovery ↗
Meta (Superintelligence Labs and FAIR) · Research scientists and engineers · PhD required
- AI Research Scientist, CoreML - Monetization ↗
Meta (Superintelligence Labs and FAIR) · Research scientists and engineers · PhD required
- AI Research Scientist, CoreML - Monetization AI ↗
Meta (Superintelligence Labs and FAIR) · Research scientists and engineers · PhD required
- Research Engineer, Monetization AI ↗
Meta (Superintelligence Labs and FAIR) · Research scientists and engineers · PhD required
- Research Scientist, Contextual AI and Multimodal Agents ↗
Meta (Superintelligence Labs and FAIR) · Research scientists and engineers · PhD required
- AI Engineer, Product ↗
Mistral AI · Applied AI and forward-deployed · No degree stated
- Applied AI - Forward Deployed Engineer - Defense ↗
Mistral AI · Applied AI and forward-deployed · Master's or PhD required
- Applied AI Engineer, AI for Mistral ↗
Mistral AI · Applied AI and forward-deployed · No degree stated
- Applied AI Engineer, Australia ↗
Mistral AI · Applied AI and forward-deployed · No degree stated
- Applied AI Engineer, CyberSecurity ↗
Mistral AI · Applied AI and forward-deployed · No degree stated
- Applied AI Engineer, CyberSecurity - Montreal ↗
Mistral AI · Applied AI and forward-deployed · No degree stated
- Applied AI Engineer, Prototyping ↗
Mistral AI · Applied AI and forward-deployed · No degree stated
- Applied AI, Forward Deployed Machine Learning Engineer ↗
Mistral AI · Applied AI and forward-deployed · Master's or PhD required
- Applied AI, Forward Deployed Machine Learning Engineer ↗
Mistral AI · Applied AI and forward-deployed · No degree stated
- Applied AI, Forward Deployed Machine Learning Engineer ↗
Mistral AI · Applied AI and forward-deployed · Master's or PhD required
- Applied AI, Forward Deployed Machine Learning Engineer - EMEA ↗
Mistral AI · Applied AI and forward-deployed · No degree stated
- Applied AI, Machine Learning Engineer, South Korea ↗
Mistral AI · Applied AI and forward-deployed · Master's or PhD required
- Applied AI, Technical Lead - Forward Deployed AI Engineer ↗
Mistral AI · Applied AI and forward-deployed · Master's or PhD required
- Applied AI, Technical Lead, Forward Deployed AI Engineer ↗
Mistral AI · Applied AI and forward-deployed · Master's or PhD required
- Applied AI, Technical Lead, Forward Deployed AI Engineer ↗
Mistral AI · Applied AI and forward-deployed · Master's or PhD required
- Applied AI, Technical Lead, Forward Deployed AI Engineer ↗
Mistral AI · Applied AI and forward-deployed · Master's or PhD required
- Applied AI, Technical Lead, Forward Deployed AI Engineer - EMEA ↗
Mistral AI · Applied AI and forward-deployed · Master's or PhD required
- Applied Scientist ↗
Mistral AI · Research scientists and engineers · PhD preferred
- Applied Scientist (Internship) ↗
Mistral AI · Fellowships, residencies and internships · PhD preferred
- Applied Scientist (Internship in Paris or London) ↗
Mistral AI · Fellowships, residencies and internships · PhD preferred
- Applied Scientist, EMEA ↗
Mistral AI · Research scientists and engineers · PhD preferred
- Applied Scientist, South Korea ↗
Mistral AI · Research scientists and engineers · PhD preferred
- Lead Applied AI, Fullstack Software Engineer, South Korea ↗
Mistral AI · Applied AI and forward-deployed · No degree stated
- Lead Applied AI, Machine Learning Engineer, South Korea ↗
Mistral AI · Applied AI and forward-deployed · Master's or PhD required
- Applied AI Architect ↗
OpenAI · Applied AI and forward-deployed · No degree stated
- Applied AI Architect ↗
OpenAI · Applied AI and forward-deployed · No degree stated · $157k to $276k base
