Databricks Unveils New Mosaic AI Capabilities to Enhance Production-Quality AI Systems

Databricks, a prominent data and AI firm, has announced several important updates to its Mosaic AI platform, targeted at assisting clients in developing production-quality Generative AI applications. This strategic investment focuses on three main areas: building compound AI systems, improving model quality, and providing new AI governance tools. These developments are intended to help organisations shift Generative AI projects from pilot to full-scale production while addressing typical issues such as privacy, quality, and cost.

Matei Zaharia, Co-founder and CTO at Databricks, emphasized the importance of these developments, stating, “We believe that compound AI systems will be the best way to maximize the quality, reliability, and measurement of AI applications going forward, and may be one of the most important trends in AI in 2024. Databricks is uniquely positioned to capitalize on these trends with the investments we’re making to improve quality, augmenting the model with real-time data and agents and tools to give it new capabilities it has little knowledge of.”

Building Compound AI Systems

Databricks released the Mosaic AI Agent Framework and the Mosaic AI Tools Catalogue to help organisations develop enterprise-ready compound AI systems. Compound AI systems use a variety of components, including models, retrievers, vector databases, and tools for assessment, monitoring, security, and governance. This method tries to create AI applications that are more accurate, safe, and controlled.

The Mosaic AI Agent Framework enables developers to swiftly and safely create high-quality Retrieval-Augmented Generation (RAG) applications based on foundation models and corporate data. Mosaic AI Agent Evaluation, an AI-assisted technique for analysing output quality and receiving input from human stakeholders, is used to evaluate these apps.

Furthermore, organisations may securely regulate, distribute, and register tools using the Mosaic AI Tools Catalogue, which is connected with Databricks Unity Catalogue. This guarantees that tool-enabled models may access these resources in a secure and regulated way, hence increasing their capabilities.

Enhancing Model Quality and Reducing Costs

Mosaic AI Model Training fine-tunes open source foundation models with an organization’s proprietary data to improve accuracy in certain domains or jobs. This fine-tuning results in smaller, more efficient models that serve faster and at a lower cost, deliv

AI Governance and Monitoring

The Mosaic AI Gateway provides a uniform interface for querying, managing, and deploying any open-source or proprietary model. It provides use tracking, safeguards, and filtering for safety and personally identifiable information (PII), guaranteeing strong governance and ongoing quality monitoring across all Generative AI applications.

Industry Impact

Several industry leaders have already benefited from Databricks’ innovations. Denis Kamotsky, Principal Software Engineer at Corning, shared, “We built an AI research assistant using Databricks Mosaic AI Agent Framework to index hundreds of thousands of documents including US patent office data. Having our LLM-powered assistant respond to questions with high accuracy was extremely important to us — that way, our researchers could find and further the tasks they were working on. To implement this, we used Databricks Mosaic AI Agent Framework to build a Generative AI solution augmented with the US patent office data. By leveraging the Databricks Data Intelligence Platform, we significantly improved retrieval speed, response quality, and accuracy.”

Tom Thomas, VP of Analytics at FordDirect, highlighted their success, stating, “FordDirect is on the leading edge of the digital transformation of the automotive industry. We are the data hub for Ford and Lincoln dealerships, and we needed to create a unified chatbot to help our dealers assess their performance, inventory, trends, and customer engagement metrics. Databricks Mosaic AI Agent Framework allowed us to integrate our proprietary data and documentation into our Generative AI solution that uses RAG. The integration of Mosaic AI with Databricks Delta Tables and Unity Catalog made it seamless to our vector indexes real-time as our source data is updated, without needing to touch our deployed model.”

Kenan Colson, VP Data & AI at Lippert, also praised the advancements, saying, “As a leading global manufacturer, Lippert leverages data and AI to build highly-engineered products, customized solutions, and the best possible experiences. Mosaic AI Agent Framework has been a game-changer for us because it allowed us to evaluate the results of our GenAI applications and demonstrate the accuracy of our outputs while maintaining complete control over our data sources. Thanks to the Databricks Data Intelligence Platform, I’m confident in deploying to production.”

Databricks had made headlines in 2023 with its $1.3 billion acquisition of MosaicML. Ghodsi had reaffirmed Databricks’ position as the premier vendor for businesses looking to train LLMs or AI on their proprietary data, adding, “When it comes to Gen AI, our strategy is very simple: We are the best vendor out there.”