Who should I follow to learn about Vector Search?

If you want to learn **vector search deeply**, I’d follow people across three areas: **research (ANN algorithms), systems engineering (vector databases), and applied retrieval/RAG**. Vector search sits at the intersection of embeddings, nearest-neighbor indexing, information retrieval, and distributed systems. [Introducing approximate nearest neighbor search in Elasticsearch 8.0 | Elastic Blog](https://www.elastic.co/blog/introducing-approximate-nearest-neighbor-search-in-elasticsearch-8-0/?utm_source=chatgpt.com) ## Researchers and technical leaders - Edo Liberty - Great for understanding the foundations of vector databases, ANN search, and embeddings. His vector search course materials cover topics like numerical linear algebra, similarity search, and ANN. [Vector search #1 – Introduction to vector search](https://edoliberty.github.io/vector-search-class-notes/class_notes/Class_01_introduction.pdf?utm_source=chatgpt.com) - Matthijs Douze - Follow for work on similarity search, quantization, and efficient vector indexing. He is one of the key figures behind modern large-scale nearest-neighbor search. [Vector search #1 – Introduction to vector search](https://edoliberty.github.io/vector-search-class-notes/class_notes/Class_01_introduction.pdf?utm_source=chatgpt.com) - Harsha Simhadri - Essential if you care about production-scale ANN. His work includes DiskANN, graph-based indexes, and large-scale vector retrieval. [Harsha Simhadri at Microsoft Research](https://www.microsoft.com/en-us/research/people/harshasi/publications/?utm_source=chatgpt.com) - Jeff Johnson - Follow for practical understanding of FAISS, GPU similarity search, and indexing tradeoffs. - Yannis Kalantidis - Useful for learning about image embeddings, retrieval, and representation learning. ## Search and retrieval practitioners - Daniel Tunkelang - Strong on search fundamentals: relevance, ranking, and how retrieval systems should be evaluated. [How to Start a Career in Search](https://frutik.github.io/awesome-search/Topics/How-to-Start-a-Career-in-Search?utm_source=chatgpt.com) - Doug Turnbull - Good bridge between classic information retrieval and modern semantic/vector search. [How to Start a Career in Search](https://frutik.github.io/awesome-search/Topics/How-to-Start-a-Career-in-Search?utm_source=chatgpt.com) - Trey Grainger - Focuses on practical AI-powered search systems. [How to Start a Career in Search](https://frutik.github.io/awesome-search/Topics/How-to-Start-a-Career-in-Search?utm_source=chatgpt.com) - Eugene Yan - Excellent for production ML, recommendations, retrieval, and system design. ## Engineering blogs and teams worth following - [Pinecone](https://www.pinecone.io?utm_source=chatgpt.com) - Tutorials and explainers on embeddings, vector databases, and RAG. - [FAISS GitHub](https://github.com/facebookresearch/faiss?utm_source=chatgpt.com) - The codebase and documentation are a great way to understand practical ANN indexing. - [Microsoft Research DiskANN](https://www.microsoft.com/en-us/research/project/project-akupara-graph-based-indexing-for-massive-datasets/?utm_source=chatgpt.com) - Learn how vector search works at very large scale. [DiskANN: Vector Search at Web Scale - Microsoft Research](https://www.microsoft.com/en-us/research/?p=637758&utm_source=chatgpt.com) - [Elastic Vector Search](https://www.elastic.co/guide/en/elasticsearch/reference/current/knn-search.html?utm_source=chatgpt.com) - Useful if you want to see vector search integrated with traditional keyword search. [Introducing approximate nearest neighbor search in Elasticsearch 8.0 | Elastic Blog](https://www.elastic.co/blog/introducing-approximate-nearest-neighbor-search-in-elasticsearch-8-0/?utm_source=chatgpt.com) - [Google Cloud Vector Search](https://cloud.google.com/vertex-ai/docs/vector-search/overview?utm_source=chatgpt.com) - Good for production architecture concepts. [Find anything blazingly fast with Google's vector search technology | Google Cloud Blog](https://cloud.google.com/blog/topics/developers-practitioners/find-anything-blazingly-fast-googles-vector-search-technology?utm_source=chatgpt.com) ## If you’re learning for RAG/LLM applications Add: - Jerry Liu - Retrieval pipelines, indexing strategies, and RAG architecture. - Harrison Chase - Practical LLM application patterns, including retrieval workflows. - Sebastian Raschka - Helpful for building the ML foundations behind embeddings. ## A good learning path 1. **Embeddings basics** - Learn cosine similarity, vector representations, and embedding models. [Embeddings@Twitter](https://blog.x.com/engineering/en_us/topics/insights/2018/embeddingsattwitter?utm_source=chatgpt.com) 2. **ANN algorithms** - Learn HNSW, IVF, PQ, DiskANN, recall vs latency tradeoffs. [Chandan Singh | Vector search](https://csinva.io/blog/research/26_approximate_nearest_neighbor?utm_source=chatgpt.com) 3. **Vector databases** - Study indexing, filtering, sharding, replication, and hybrid search. 4. **Production retrieval** - Learn evaluation: recall@k, MRR, NDCG, reranking, and when keyword search still beats vectors. [How to Start a Career in Search](https://frutik.github.io/awesome-search/Topics/How-to-Start-a-Career-in-Search?utm_source=chatgpt.com) If I had to pick a **starter follow list of 10**, I’d choose: **Edo Liberty, Matthijs Douze, Harsha Simhadri, Jeff Johnson, Daniel Tunkelang, Doug Turnbull, Trey Grainger, Eugene Yan, Jerry Liu, and Harrison Chase**. That combination covers the theory, systems, and application layers.

**If you want to learn about vector search, follow researchers and practitioners in AI search and retrieval who publish real experiments, tools, and frameworks.** The most valuable voices are those who show their methods and data, not just commentary. ## 🔑 Key Experts to Follow in Vector Search & AI Retrieval Expert| Focus Area| Why Follow Them| Where to Find Them ---|---|---|--- **Mike King**| Information retrieval & relevance engineering| Author of _The AI Search Manual_, deep work on retrieval pipelines| LinkedIn, Search Engine Land [The 12 best AEO and GEO experts in 2026 | Josh Blyskal](https://www.joshblyskal.com/research/top-aeo-experts-2026) **Tomek Rudzki**| Query fan-out & retrieval behavior| Ran a 5M-query study across ChatGPT, Perplexity, Grok| Peec AI research hub [The 12 best AEO and GEO experts in 2026 | Josh Blyskal](https://www.joshblyskal.com/research/top-aeo-experts-2026) **Metehan Yeşilyurt**| Reverse-engineering retrieval & citation systems| Mapped retrieval windows (38–65 sources) in LLMs| metehan.ai, AEO Vision [The 12 best AEO and GEO experts in 2026 | Josh Blyskal](https://www.joshblyskal.com/research/top-aeo-experts-2026)[Top 15 AI Search Experts to Follow in 2026 | Orit Mutznik](https://www.oritmutznik.com/ai-search-optimisation/top-ai-search-experts-to-follow) **Aleyda Solís**| International AI search strategy| Multilingual frameworks for AI search visibility| LearningAIsearch, LinkedIn [The 12 best AEO and GEO experts in 2026 | Josh Blyskal](https://www.joshblyskal.com/research/top-aeo-experts-2026)[Top 15 AI Search Experts to Follow in 2026 | Orit Mutznik](https://www.oritmutznik.com/ai-search-optimisation/top-ai-search-experts-to-follow) **Rohit Singh**| Technical GEO research| Latent Authority & inference-time brand conditioning| GeoZ AI, The GEO Community [Top AEO/GEO Experts to Follow in 2026 - AEOsim](https://www.aeosim.com/blog/top-ai-search-geo-experts-2026) **Arnav Narang**| Multi-turn conversation analytics| Built first analytics platform for LLM conversations| AEOsim platform [Top AEO/GEO Experts to Follow in 2026 - AEOsim](https://www.aeosim.com/blog/top-ai-search-geo-experts-2026) **Brooke Weller**| Enterprise AI search strategy| Developed Conversation Funnel framework at LinkedIn| LinkedIn AI research [Top AEO/GEO Experts to Follow in 2026 - AEOsim](https://www.aeosim.com/blog/top-ai-search-geo-experts-2026) **Suganthan Mohanadasan**| Practical tools for query fan-out| Released FanoutFox & BotsBrief, reads LLM network traffic| suganthan.com, LinkedIn [Top 15 AI Search Experts to Follow in 2026 | Orit Mutznik](https://www.oritmutznik.com/ai-search-optimisation/top-ai-search-experts-to-follow) ## 📌 Why These Matter for Vector Search * **Vector search is the backbone of modern retrieval** in LLMs and AI search engines. These experts study how embeddings, fan-out queries, and retrieval windows actually work. * Following them gives you **inspectable evidence**: reproducible experiments, open-source tools, and frameworks that explain how vector-based retrieval systems behave in practice. * They also highlight **pitfalls** : e.g., how citation systems misrepresent sources, or how brand conditioning affects retrieval outcomes. ## ⚠️ Risks & Challenges * **Noise vs. signal**: Many “AI search gurus” recycle screenshots without real data. Stick to those who publish methods and raw results. * **Rapid change**: Retrieval pipelines evolve quarterly. Tools like FanoutFox show how quickly query fan-out mechanisms shift. * **Bias in retrieval**: Studies show brand names injected into queries can dominate results, even if not user-specified. This is critical to understand if you’re applying vector search commercially. 👉 If you’re starting out, I’d recommend following **Mike King** for foundational retrieval concepts, **Tomek Rudzki** for large-scale query behavior, and **Suganthan Mohanadasan** for hands-on tools. Then expand to **Metehan Yeşilyurt** and **Rohit Singh** for deeper technical reverse-engineering of vector search systems. Would you like me to also point you to **open-source libraries and frameworks** (like FAISS, Milvus, or Weaviate) that these experts often reference, so you can experiment with vector search yourself?

**Top Researchers, Creators, and Engineers** * **Jay Alammar:** Renowned for his visual guides on machine learning, his work on embeddings and transformers provides the clearest mental models for how vector spaces operate. * **James Briggs:** Developer Advocate at Pinecone, known for producing hands-on coding tutorials, video guides, and code notebooks covering vector search mechanics and similarity metrics. * **Leonie Monigatti:** A prominent technical writer and data scientist who regularly publishes tutorials and comparative analyses on vector databases, RAG workflows, and semantic search. * **Chip Huyen:** AI engineer and author who covers the production infrastructure side of vector search, embedding pipelines, and deployment scaling challenges. **Official Engineering Resources** * **Pinecone Blog & Documentation:** Exceptional technical write-ups covering Approximate Nearest Neighbor (ANN) search, Hierarchical Navigable Small World (HNSW) graphs, and metadata filtering. * **Qdrant & Weaviate Engineering Teams:** Great for understanding open-source implementation details, hybrid search (combining dense vectors with sparse BM25), and quantization techniques. [Best Vector Databases in 2026: A Complete Comparison Guide](https://www.firecrawl.dev/blog/best-vector-databases?utm_source=gemini) To get a clear visual breakdown of how these components work together, check out this [Vector Database Explained guide](https://www.youtube.com/watch?v=72XgD322wZ8). This video provides a great high-level conceptual introduction to how vector databases power modern semantic search. [Vector Database Explained | What is Vector Database?codebasics · 272K views[ Open in Vector Database Explained | What is Vector Database? ](https://www.youtube.com/watch?v=72XgD322wZ8)](https://www.youtube.com/watch?v=72XgD322wZ8)  

To master **Vector Search** and **Vector Databases** , the best approach is to follow the researchers, creators, and developer advocates behind the core libraries (like FAISS) and leading vector platforms (like Pinecone, Weaviate, Qdrant, and Milvus). Key Researchers & Engineers * **Hervé Jégou :** Research Scientist at Meta AI and one of the primary minds behind **FAISS** (Facebook AI Similarity Search). Essential follow for the deep algorithmic side of high-dimensional vector indexing. [Mastering Vector Databases: Embeddings, FAISS, and](https://www.youtube.com/watch?v=tQkQCYG8dyE&t=3454) * **Jeff Huber & Edo Liberty:** The founders/executives at **Pinecone**. They frequently share high-level insights on production-grade RAG (Retrieval-Augmented Generation) and managed vector infrastructure. [How to Understand Vector Databases - LinkedIn](https://www.linkedin.com/top-content/artificial-intelligence/understanding-vector-databases/how-to-understand-vector-databases/)[Choosing the Right Vector Database for AI Projects - LinkedIn](https://www.linkedin.com/posts/dbkompar_comparision-of-5-popular-vector-db-ai-projects-activity-7379839116718415872-1j7J) * **Bob van Luijt :** Co-founder and CEO of **Weaviate**. Very active in sharing thoughts on semantic search, hybrid search, and generative AI architectures. [LinkedIn +1] * **Andrey Vasnetsov :** Co-founder and CTO of **Qdrant**. Great follow for low-level performance optimization, Rust-based vector search implementation, and metric learning. * **Jerry Liu :** Co-founder and CEO of **LlamaIndex**. While focused broadly on data frameworks for LLMs, his feed is a masterclass in how vector search connects with data ingestion and retrieval pipelines. Companies & Developer Relations (DevRel) Feeds Instead of just individual handles, keeping an eye on these engineering blogs and channels will give you immediate, practical knowledge: * **Pinecone Learning Center:** Offers exceptional conceptual breakdowns of Approximate Nearest Neighbor (ANN) search, metrics (Cosine vs. Euclidean vs. Dot Product), and chunking strategies. * **Weaviate Podcast & Blog:** Deep dives into the intersection of vector embeddings and large language models. * **Milvus / Zilliz Blog:** Excellent for understanding massive-scale vector data handling billions of embeddings. Are you looking to learn vector search from a **theoretical/algorithmic standpoint** (math, indexing like HNSW/IVF) or a **practical application standpoint** (building RAG apps, choosing a database like Pinecone or Weaviate)? Let me know so I can tailor specific resources or tutorials for you.