MongoDB · Atlas AI Search AI search on MongoDB Atlas: vector, full-text and hybrid.
We engineer the retrieval layer for retrieval-augmented generation (RAG) and agents on MongoDB Atlas — Atlas Vector Search, Atlas Search, hybrid retrieval, and Voyage AI embeddings and reranking, all on your operational data. One database for documents and vectors, with no separate vector store to run or sync.
Why it matters
Most RAG projects stall on retrieval, not the model.
The model is the easy part. The retrieval layer is where accuracy is won or lost.
- 01Bolt-on vector stores add a second system to provision, a sync pipeline to maintain and a whole class of dual-write and staleness bugs.
- 02Pure semantic search misses queries where exact terms carry the meaning — a code identifier, a policy number, a proper noun.
- 03Pure keyword search misses queries where intent matters more than the words used.
- 04Answers without citations cannot be reviewed, audited or held to account.
What we engineer
One platform for operational data and vectors.
Embeddings, documents, full-text indexes and metadata live together — queried through one API and governed by one access model.
Atlas Vector Search
Semantic similarity search over vector embeddings that finds the documents closest in meaning to the query. It powers RAG, agentic retrieval and semantic search over products, documents, tickets and knowledge bases.
Hybrid search
Atlas Search provides Lucene-based full-text search with BM25-style relevance; Atlas Vector Search provides the semantic half. Fused together, both kinds of query land the right documents — our default for production RAG.
Voyage AI embeddings
The right Voyage AI model for each corpus and language — general-purpose, contextualised chunk, multimodal, code, finance and legal — with quantised embeddings where storage and latency costs matter.
Reranking
A Voyage AI reranker (rerank-2.5) re-scores candidates by true query relevance — multilingual, over embedding or lexical results — and lifts the best to the top before generation.
Citation-grounded answers
Every generated answer carries its sources, so responses are traceable and auditable rather than free-floating.
No sync layer
When a document changes, the retrieval index sits on the same data — not a downstream copy you have to reconcile. Atlas runs on AWS, Azure and Google Cloud, in the same cloud and region as your stack.
How we build RAG on Atlas
The retrieval layer, built and evaluated in full.
/ 01
Chunking
Split source documents into retrieval units sized for the corpus and the questions users actually ask — not arbitrary fixed windows.
/ 02
Embedding
Choose the right Voyage AI model for each corpus (general voyage-4-large, contextual voyage-context-3, code, multimodal, finance, legal), with quantised embeddings where cost and latency matter.
/ 03
Hybrid retrieval
Combine Atlas Vector Search and Atlas Search so both semantic and exact-term queries land the right documents.
/ 04
Reranking
Apply a Voyage AI reranker (rerank-2.5) to the candidate set to lift the most relevant results to the top before generation.
/ 05
Citation-grounded answers
Every generated answer carries its sources, so responses are traceable and auditable.
/ 06
Evaluation
Measure retrieval and answer quality with Ragas and custom scorers from our Agent Evaluations practice — on your corpus and queries, not vendor benchmarks — so every change is proven, not assumed.
The stack
One coherent retrieval stack on your operational data.
Database and vectors
- MongoDB Atlas
- AWS
- Azure
- Google Cloud
Retrieval
- Atlas Vector Search
- Atlas Search (Lucene)
- Hybrid retrieval
Embeddings and reranking
- voyage-4-large
- voyage-context-3
- voyage-multimodal-3.5
- voyage-code-3
- Finance and legal models
- rerank-2.5
- rerank-2.5-lite
Orchestration
- LangChain
- LlamaIndex
- CrewAI
Evaluation
- Ragas
- Custom scorers
- Agent Evaluations
Use cases
- RAG applications
- Fraud prevention
- Claims-processing modernisation
- Code search
- Legal and financial retrieval
Voyage AI models
Chosen per corpus and language, with reranking on top.
voyage-4-large
- What it is
- Top general-purpose and multilingual retrieval quality
- When we use it
- The default when retrieval accuracy is paramount; the voyage-4 series shares one embedding space
voyage-context-3
- What it is
- Contextualised chunk embeddings
- When we use it
- Long, chunked documents — each chunk embedded with its surrounding context; multilingual
voyage-multimodal-3.5
- What it is
- Rich multimodal embeddings
- When we use it
- Interleaved text and visuals — PDF screenshots, slides, tables, figures and video
voyage-code-3
- What it is
- Code retrieval with lower-dimensional quantised embeddings
- When we use it
- Code search
rerank-2.5
- What it is
- Generalist, instruction-following reranker
- When we use it
- Re-scoring candidates over embedding or lexical (BM25/TF-IDF) results; multilingual
rerank-2.5-lite
- What it is
- Lighter reranker
- When we use it
- Where a lighter reranking pass is enough
What you get
Governance, residency and evaluation, built in.
Hybrid retrieval on the data you already run — no second system
India data residency — retrieval in an Indian region alongside your primary data
DPDP-aware governance — access controls and provenance metadata on retrieved chunks
Deletion that reaches the embeddings, not just the primary documents
Retrieval quality measured over time, so regressions surface before users notice
Citation-grounded answers that make review and audit practical
Where it fits
The retrieval layer under your RAG and agents.
Atlas retrieval pairs with our RAG practice and with the Data Platform that holds your operational data, and it relates to InVocIQ, our voice and natural-language BI solution. Our Agent Evaluations practice keeps retrieval quality measured on your corpus and queries.
FAQ
Frequently asked.
01What is MongoDB Atlas Vector Search?
Semantic similarity search over vector embeddings stored alongside your operational data. It finds the documents closest in meaning to a query — the retrieval layer that powers RAG, agentic retrieval and semantic search. Because vectors live in the same database as your documents, there is no separate vector store to run or sync.
02What is hybrid search?
Hybrid search combines Atlas Vector Search (semantic) with Atlas Search (Lucene-based full-text, BM25 lexical matching) and fuses the results. It catches both the queries where meaning matters and the queries where exact terms matter. On most enterprise corpora, it is more relevant than either approach alone.
03What is Voyage AI?
MongoDB’s family of embedding and reranking models. Featured embeddings include voyage-4-large (general-purpose, multilingual), voyage-context-3 (contextualised chunks) and voyage-multimodal-3.5 (text and visuals, including video); voyage-code-3 and finance and legal models extend the family. The rerankers — rerank-2.5 (generalist, instruction-following) and the lighter rerank-2.5-lite — re-score query–document relevance over embedding or lexical (BM25/TF-IDF) results.
04Do I need a separate vector database?
Not with Atlas. It unifies operational data and vector embeddings in one database, so there is no separate vector store to provision, operate or synchronise. That removes dual-write and staleness bugs and simplifies governance, backup and access control.
Related guides
Keep reading.
Data Engineering · 8 min
Vector Database for the Enterprise.
Pinecone, Milvus, Weaviate, Qdrant, pgvector, OpenSearch — the trade-offs that actually matter for production RAG in India.
Read the guide
AI Engineering · 6 min
RAG vs Fine-Tuning
When to use retrieval augmented generation versus fine-tuning for enterprise LLM applications.
Read the guide
AI Consulting · 9 min
DPDP Act AI Compliance, India.
The 2026 enterprise playbook for the Digital Personal Data Protection Act — phased timeline, automated-decision rights, and the controls every AI system needs.
Read the guideNext step
Win accuracy where it is won: retrieval.
Semantic, lexical and hybrid search on the data you already run, Voyage AI embeddings and reranking, and every answer grounded in citations — with no second system to provision and no sync pipeline to maintain.