Engineering teams globally are developing AI-centric applications or integrating AI functions into existing products, spurred by a maturing AI development ecosystem that facilitates rapid prototyping. Yet, transitioning AI applications to production remains notoriously complex, with modern AI stacks demanding LLMs, embeddings, vector search, observability, new caching layers, and continual adaptation to an ever-changing landscape. The data layer increasingly represents both the foundation and bottleneck in AI app production.
MongoDB is expanding its core document database into a comprehensive AI-ready platform with capabilities in operational data, search, real-time analytics, and AI-driven data retrieval. The company recently acquired Voyage AI to provide precise and cost-effective embedding models and rerankers.
Fred Roma, an experienced engineer and the SVP of Product and Engineering at MongoDB, joins the show with Kevin Ball to discuss AI application development, vector search and reranking, schema evolution in the LLM era, the Voyage AI acquisition, and how data platforms must adapt to AI’s rapid evolution.
Full Disclosure: This episode is sponsored by MongoDB.
Kevin Ball, known as KBall, is the vice president of engineering at Mento and an independent coach for engineers and engineering leaders. He co-founded and was CTO of two companies, founded the San Diego JavaScript meetup, and organizes the AI inaction discussion group through Latent Space.
Please click here to see the transcript of this episode.