MongoDB announced new capabilities at its MongoDB.local Bengaluru event on June 30, 2026, that address the two biggest reasons enterprise AI projects stall before production: retrieval that isn't accurate enough to trust and infrastructure that can't meet compliance requirements. The company launched Native Reranking, Hybrid Search, and a new embedding model-all powered by Voyage AI models that currently lead the public Retrieval Embedding Benchmark leaderboard-and brought its Search and Vector Search features to on-premises and self-managed environments for the first time.
"The biggest barrier to enterprise AI in production and at scale isn't the LLM. It's memory, retrieval, accuracy, and compliance," said Ben Cefalo, Chief Product Officer, Core Products at MongoDB. "Most enterprises aren't blocked by ambition. They're held back by infrastructure that wasn't designed to provide AI with trusted access to enterprise data. Whether you're running in the cloud, private cloud, or behind a firewall, MongoDB gives you the same production-grade retrieval capabilities wherever your data lives."
Three new capabilities built into the database
Native Reranking in MongoDB Atlas, now in public preview, runs inside the aggregation pipeline without external APIs and delivers up to a 30% boost in retrieval quality directly inside the database. Voyage Context 4, generally available, is an embedding model built for long documents that processes content in full context rather than isolated chunks, preserving meaning across complex enterprise content. Hybrid Search, also generally available, combines full-text and vector search in a single query on live operational data, so agents retrieve from the current state of the data rather than a stale copy.
Mukund Jha, CEO of Emergent Labs, an AI-native app development platform, said: "Our agents write code, modify data structures, and act on what they read back millions of times a day. If retrieval returns something stale or wrong, the agent builds on it, and the error compounds. MongoDB gives us the retrieval accuracy to keep agents working from the current state of the data, and that's what lets us run two million applications at scale."
Bringing AI retrieval to where enterprise data lives
Search and Vector Search are now generally available for MongoDB Enterprise Advanced, letting organizations run AI retrieval on-premises, in private clouds, or hybrid environments under their own compliance frameworks. More than 20 of the world's largest banks and financial institutions have been evaluating the feature ahead of this release. The same capabilities are also available in MongoDB Community Edition at no cost, so startups can prototype locally and then scale to Atlas or Enterprise Advanced without re-architecting.
Investing in India's builder ecosystem
As part of the event, MongoDB announced plans to upskill two million Indian builders by 2030 through partnerships with the All India Council for Technical Education, HCL GUVI, and the ICT Academy of Kerala. The company also launched a startup challenge called Bengaluru to the Bay, offering $50,000 in MongoDB Atlas credits, travel, and access to San Francisco's AI community for winning founders.
Why this matters for IT and development professionals
For teams building AI applications, retrieval accuracy and data compliance are not optional. MongoDB's new stack means developers can implement hybrid search and reranking directly inside the operational database, eliminating the need for separate vector databases or external APIs. The availability of these features in Enterprise Advanced and Community Edition gives IT architects a single platform that spans development laptops, private clouds, and regulated on-premises environments. The up to 30% retrieval quality improvement from Native Reranking directly reduces the risk of AI agents acting on incorrect data, a failure mode that compounds quickly in production systems. For professionals focused on AI for IT & Development, these updates provide a production-ready retrieval stack that is accurate, compliant, and deployable wherever their data lives.
Your membership also unlocks:
