Approximate Nearest Neighbor Search Courses and Certifications

Coursera

Master ANN Search

Build vector indexes with FAISS and Annoy, benchmark recall–latency trade-offs against brute-force search, and tune index parameters to optimize a 100k-vector dataset.

• Coursera
• 3 weeks, 1 hour a week
• Self-Paced
• Paid Course

James Briggs

HNSW for Vector Search Explained and Implemented with Faiss - Python

Understand HNSW's layered proximity graphs for approximate nearest-neighbor search, then implement and tune an HNSW index with Faiss.

• YouTube
• 35 minutes
• Self-Paced
• Free Video

Pinecone

Graph-Based Approximate Nearest Neighbors and HNSW

Explore graph-based approximate nearest neighbor search, with a practical focus on HNSW construction, memory constraints, updates, deletions, and benchmarking.

• YouTube
• 59 minutes
• Self-Paced
• Free Video

Neural Search Improvements with Apache Solr 9.1 - Approximate Nearest Neighbor and Pre-Filtering

Explore neural search in Apache Solr 9.1, including HNSW graph improvements, pre-filtering capabilities, and Deep Learning integrations for enhanced search relevance and efficiency.

• YouTube
• 39 minutes
• Self-Paced
• Free Video

Optimizing and Scaling ChromaDB for Vector Search

Optimize ChromaDB for large-scale vector search by reducing retrieval latency, narrowing search spaces, and processing real-time data streams.

• CodeSignal
• 2 hours
• Self-Paced
• Free Certificate

Optimizing and Scaling Pinecone for Vector Search

Scale Pinecone vector search by reducing query latency, narrowing dynamic search spaces, and monitoring real-time stream performance.

• CodeSignal
• 1 hour
• Self-Paced
• Free Certificate

Indexing, Optimization and Scaling pgvector

Optimize and scale pgvector queries through indexing, performance monitoring, and tuning IVFFlat and HNSW search parameters.

• CodeSignal
• 1 hour
• Self-Paced
• Free Certificate

Coursera Career Certificate

Vector DB Foundations, Embeddings & Search Algorithms

Build semantic search systems: generate text and image embeddings, tune HNSW parameters, benchmark FAISS and Annoy ANN indices, and implement RAG and multimodal search with Weaviate.

• Coursera
• 4 weeks, 10 hours a week
• Self-Paced
• Paid Course
• 8 courses

Retrieval Optimization: Tokenization to Vector Quantization

Optimize vector search for RAG: compare Byte-Pair Encoding, WordPiece, Unigram and SentencePiece tokenizers, tune HNSW parameters, and apply product, scalar and binary quantization.

• Coursera
• 1 hour 30 minutes
• Self-Paced
• Free Online Course

Coursera

Vector Database Foundations and Core Concepts

Build vector search systems: generate embeddings from text and images, compare cosine, dot-product and Euclidean metrics, implement FAISS and Annoy, and tune HNSW parameters.

• Coursera
• 6 weeks, 2 hours a week
• Self-Paced
• Paid Course

Coursera

Optimize SQL and Vector Search Parameters

Tune parameterized SQL queries and ANN vector search parameters with SQLite, FAISS, and Annoy, benchmarking recall, latency, throughput, and resource use in retrieval systems.

• Coursera
• 3 weeks, 2 hours a week
• Self-Paced
• Paid Course

Coursera

Unify Modalities: Cross-Modal Retrieval

Build cross-modal retrieval systems: implement approximate nearest-neighbor search with FAISS and design attention-based fusion of visual and textual embeddings using datasets like Flickr30K.

• Coursera
• 2 weeks, 1 hour a week
• Self-Paced
• Paid Course

Candidate Generation & Retrieval Architectures

Build multi-source candidate retrieval pipelines combining collaborative, content, graph, sequential, and generative methods for large catalogs and low-latency serving.

• Coursera
• 16 weeks, 1 hour a week
• Self-Paced
• Paid Course

Career Certificate

Microsoft Recommender Systems Engineering with LinkedIn

Build production-grade recommender systems through LinkedIn-inspired architectures, from two-tower and graph retrieval to generative ranking and Azure MLOps.

• Coursera
• 4 weeks, 10 hours a week
• Self-Paced
• Paid Course

James Briggs

FAISS - Introduction to Similarity Search

Learn how FAISS indexes vector embeddings to make high-dimensional similarity search faster.

• YouTube
• 32 minutes
• Self-Paced
• Free Video