Building a Unified Data Platform for Gen AI

In today’s digital-first world, data is the lifeblood of innovation and decision-making. Yet, businesses often find themselves constrained by outdated and fragmented systems that fail to meet the demands of a fast-paced, interconnected landscape. Legacy architectures—such as the 1970s-era mainframes still used in industries like banking—create inefficiencies, siloed data, and operational bottlenecks, leaving organizations struggling to deliver timely, actionable insights.

The pressure to adapt is mounting, as customer expectations for real-time interactions and personalized services continue to grow. To thrive in this competitive environment, organizations must embrace a transformative approach to managing their data estates—one that integrates advanced technologies seamlessly.

Unified data platforms powered by operational data layers (ODLs), generative AI (gen AI), and vector search are the solution. These innovations do more than just modernize data handling; they unlock new opportunities for agility, efficiency, and value creation, empowering businesses to make informed decisions, improve customer experiences, and drive growth.

Let’s explore how these technologies are reshaping the way businesses consume, integrate, and leverage their data.

Diagram showing the conceptual model of a converged AI data store. On the left are boxes for Structured legacy data, structured data, and unstructured date. Structured legacy data connects to the converged AI data store via the MongoDB relational migrator. Structured data connects directly to the converged AI data store. Finally, unstructured data connects to the converged AI data store through the embedding model and vectors.
Figure 1. Conceptual model of a Converged AI Data Store, showing multimodal data ingest.

From stale to real-time data: The case for operational data layers

In the rapidly evolving digital landscape, businesses can no longer afford to rely on outdated, batch-processed data. The demands of modern operations require instant access to fresh, accurate information. Yet many organizations continue to struggle with fragmented systems that deliver stale data, creating roadblocks in decision-making and customer engagement.

This is where the concept of an ODL becomes transformative. Acting as a centralized hub, an ODL integrates data from multiple transactional systems in real-time, ensuring businesses have a unified and up-to-date view of their operations. Let’s explore how ODLs can revolutionize business processes:

1. Enabling real-time customer interactions

Imagine a customer service representative handling a support call. Without real-time access to the latest data—such as a customer’s recent transactions, support history, or preferences—the interaction may feel disconnected and inefficient.

An ODL solves this problem by consolidating and providing real-time data. For example, a telecom provider could use an ODL to ensure its agents have immediate access to recent billing information, technical issues reported, and ongoing resolutions. This not only empowers the agents but also leaves the customer with a seamless and satisfactory experience.

2. Streamlining account management

Real-time data isn’t just about resolving customer issues; it’s also critical for proactive engagement. In industries like banking and retail, customers often need immediate updates on their accounts, such as current balances, transaction details, or loyalty points.

By integrating APIs with the ODL, businesses can offer instantaneous responses to these queries. For instance, a retail bank could enable customers to check recent purchases or transfers through a chatbot that queries the ODL in real-time, delivering fast, accurate results.

3. Enhancing compliance and reporting

Highly regulated industries, such as finance and healthcare, face additional challenges in managing large volumes of historical data for audits and compliance. Traditional systems often struggle to handle such demands, resulting in time-consuming manual processes.

ODLs, when combined with gen AI, enable businesses to extract, summarize, and structure this data efficiently. For instance, a financial institution could use an ODL to generate compliance reports that pull data from diverse sources—both structured and unstructured—and ensure they meet regulatory standards with minimal manual intervention.

4. Supporting metadata and governance

Another often overlooked advantage of an ODL is its ability to support metadata management and data governance. For large enterprises operating across multiple geographies, changes in localized data models are frequent and complex.

An ODL can act as a centralized repository, capturing these updates and enabling advanced search functionalities for impact analysis. For example, a global enterprise could use an ODL to track changes in data definitions, understand usage patterns, and ensure compliance with governance policies across regions—all while reducing the risk of errors.

The transformative power of gen AI and vector search

As businesses transition to real-time data strategies powered by ODLs, the potential to unlock even greater insights lies in adopting cutting-edge tools like gen AI and vector search. These technologies are revolutionizing the way organizations consume and interpret data, enabling unprecedented efficiency and intelligence.

  • Gen AI: By generating actionable insights, predictions, and content, gen AI enables businesses to turn static data into a strategic resource. For example, a retailer could use gen AI to analyze customer purchase histories and recommend personalized product bundles.

  • Vector search: This technology translates high-dimensional data like text, images, and audio into vectors, enabling accurate, intuitive searches. For instance, healthcare providers can search for similar patient cases by symptoms, enhancing diagnostics and treatment planning.

By incorporating these tools into an ODL, businesses can go beyond basic data integration, creating smarter, more agile operations capable of delivering value in real-time.

Diagram showing RAG implementation, using the converged AI data store to provide context to the LLM prompt. In the top right corner is the User, and step 1 of the process is a prompt in natural language to the embedding model. Step 2 is that the vectorized prompt is used to retrieve context from the Converged AI data store. Step 3 is prompt LLM with context and original question. Finally, in step 4, the user receives the answer.
Figure 2. Retrieval Augmented Generation (RAG) implementation, using the converged AI data store to provide context to the LLM prompt.

New opportunities: Revolutionizing operations with gen AI and operational data layers

The integration of gen AI and vector search into ODLs opens up a world of opportunities for businesses to enhance customer experience, streamline operations, and innovate at scale.

Here’s how these technologies drive transformation:

  • Enhanced data discovery: With vector search, organizations can quickly and accurately retrieve relevant data from massive datasets, simplifying complex searches.

  • Improved customer experience: Gen AI–powered ODLs analyze customer behavior to deliver personalized recommendations, building stronger customer relationships.

  • Increased operational efficiency: Automating routine data tasks with gen AI reduces manual effort, enabling teams to focus on strategic initiatives.

  • Enhanced agility and innovation: By enabling rapid development of AI-driven applications, businesses can quickly adapt to market changes and stay ahead of the competition.

As organizations embrace these capabilities, they position themselves to thrive in an increasingly competitive and data-driven world.

Architectural options for data processing

Modernizing data platforms requires a robust architecture that can handle both batch and real-time processing. Depending on their needs, organizations often choose between lambda or kappa architectures, and MongoDB can serve as a flexible operational layer for both.

The lambda architecture

The lambda architecture is ideal for organizations that need to process both batch and real-time data. It consists of three layers:

  • Batch layer: This layer processes large volumes of historical data offline. Gen AI can enrich this data by generating insights and predictions.

  • Speed layer: This layer handles real-time data streams, enabling immediate responses to changes.

  • Serving layer: This layer combines batch and real-time data into a unified view, powered by MongoDB for seamless queries and data access.

The kappa architecture

For businesses focused on real-time analytics, the kappa architecture simplifies operations by using a single stream for data processing. MongoDB excels as the operational speed layer in this setup, supporting high-speed, real-time data updates enhanced by gen AI.

By choosing the right architecture and leveraging MongoDB’s capabilities, businesses can ensure their data platforms are future ready.

A journey toward data modernization

Data modernization is a progressive journey, transforming businesses step by step into smarter, more agile systems. It begins with a basic operational data store, where read-heavy workloads are offloaded from legacy systems into MongoDB, boosting performance and accessibility.

Next comes the enriched ODL, adding real-time analytics to turn raw data into actionable insights. Then, as needs grow, parallel writes enable MongoDB to handle write-heavy operations, enhancing speed and reliability.

In the transition to the system of transaction, monolithic systems are replaced with agile microservices directly connected to MongoDB, simplifying operations and accelerating innovation. Finally, businesses reach the system of record, a domain-driven architecture where MongoDB provides unmatched scalability, flexibility, and efficiency.

Each phase of this journey unlocks new opportunities, transforming data into a dynamic asset that powers innovation, operational excellence, and growth.

Figure 3. A conceptual model showcasing the joint implementation of the Kappa (Data in Motion) and Lambda (Data at Rest) frameworks on MongoDB Atlas, utilizing Stream Processing for real-time data and Online Archive/Federation features for historical data management.

The unified and intelligent future of data

As businesses embrace real-time data architectures and advanced AI capabilities, the potential for innovation is boundless. With solutions like MongoDB, organizations can seamlessly integrate and harness their data, driving operational excellence and delivering exceptional customer experiences. Now is the time to modernize, innovate, and unlock the full potential of your data.

Discover how TCS and MongoDB are harmonizing technologies for the future. Start your data modernization journey today!

← Previous

AI-Powered Retail With Together AI and MongoDB

Generative AI (gen AI) is changing retail in fascinating ways. It’s providing new avenues for IT leaders at retailers to enhance customer experiences, streamline operations, and grow revenue in a fast-paced environment. Recently, we’ve been working closely with a fascinating organization in this space—Together AI. In this blog, we’ll explore how Together AI and MongoDB Atlas tremendously accelerated the adoption of gen AI by combining the capabilities of both platforms to bring high-impact retail use cases to life. Introduction to Together AI and MongoDB Atlas From the first look, it’s impressive how well Together AI is designed for gen AI projects. It’s a powerful platform that lets developers train, fine-tune, and deploy open-source AI models with just a few lines of code. This is a critical component for retrieval-augmented generation (RAG) . With RAG, AI can pull real-time business-specific data from MongoDB Atlas , which means retailers get more reliable and relevant outputs. That’s crucial when dealing with data as dynamic as customer behavior or inventory movement from online and physical stores. With its flexible data model, MongoDB Atlas is an ideal database engine for handling diverse data needs. It’s fully managed, multi-cloud, and exceptional at managing different data types, including the vector embeddings that power AI applications. One important feature is MongoDB Atlas Vector Search , a smart library that stores and indexes vector embeddings, making it simple to integrate with Together AI. This lets retailers generate timely, personalized responses to customer queries, creating a better experience all around. Identifying retail use cases With Together AI and MongoDB Atlas working together, the possibilities for retail are huge. Here are some of the use cases we’ve been exploring and testing with clients, each bringing measurable value to the table: Product description generation Product onboarding to a retail e-commerce portal is a time-consuming effort for many retailers. They need to ensure they’ve created a product description that matches the image, then deploy it to their e-commerce portal. For multilingual portals and multiple operating geographies, this challenge of accuracy increases. With Together AI’s support for multimodal models (e.g. Llama 3.2) and MongoDB Atlas’s vector embeddings, we can create accurate product descriptions in multiple languages. Check out a demo app to see it in action. Figure 1. Demo application for generating product descriptions. Personalized product recommendations Imagine being able to offer each customer exactly what they’re looking for, without them even asking. With Together AI’s retrieval and inference endpoints and MongoDB Atlas Vector Search, we can create highly personalized product recommendations. Analyzing individual preferences, browsing history, and past purchases becomes seamless, giving customers exactly what they need, possibly exactly when they need it. Conversational AI-powered tools (a.k.a. chatbots) We’re also deploying intelligent conversational tools that can understand complex questions, offer personalized assistance, and drive conversions. Together AI, paired with MongoDB Atlas, makes these bots responsive and relevant so customers feel like they’re talking to a knowledgeable adviser rather than a chatbot. When real-time data informs the responses, customer experience is enhanced. Dynamic pricing and promotions Pricing in retail is often a moving target, and AI-driven insights help us optimize our approach. We’ve used Together AI and MongoDB Atlas to analyze market trends, competitor pricing, and customer demand to keep our pricing competitive and adjust promotions in real-time. It’s incredible how much more strategic we can be with AI’s help. Inventory management and forecasting This might be one of the most impactful use cases I’ve worked on—using AI to predict demand and optimize stock levels. With Together AI and MongoDB Atlas, it’s easier to balance inventory, reduce waste, and ensure the products customers want are always in stock. This leads to better efficiency and fewer out-of-stock scenarios. Implementing retail use cases with Together AI and MongoDB Atlas Let me share a concrete example that really brings these concepts to life. Case study: Building a multilingual product-description-generation system We recently worked on a solution to create a product-description-generation system for an e-commerce platform. The goal was to provide highly descriptive product information based on the images of the products from the product catalog. This use case really demonstrated the value of storing the data in MongoDB and using the multilanguage capabilities of Together AI’s inference engine. Embeddings and inference with Together AI: Together AI generated product descriptions based on images retrieved from the product catalog using Llama 3.2 vision models. This way, each product’s unique characteristics were considered, then generated in multiple languages. These descriptions could then be embedded into the MongoDB Atlas Vector Search database via a simple API. Indexed embeddings with MongoDB Atlas Vector Search: Using MongoDB Atlas Vector Search capabilities, we created embeddings, and then indexed them to be used to retrieve relevant data based on other matched product queries. This step made sure the product descriptions were not just accurate but also relevant to the images. Real-time data processing: By connecting this setup to a real-time product dataset, we ensured that product descriptions in multiple languages were always updated automatically. So when a marketplace vendor or retailer uploads new images with distinct characteristics, they get up-to-date product descriptions in the catalog. This project showcased how Together AI and MongoDB Atlas could work together to deliver a solution that was reliable, highly efficient, and scalable. The feedback from users was overwhelmingly positive. They especially appreciated how intuitive and helpful the product descriptions were and how simple the whole product onboarding process could become for multilingual businesses spread across multiple geographical regions. Figure 2. An example of a query and response flow for a RAG architecture using MongoDB and Together AI. Looking at the business impacts For a retail organization, implementing Together AI and MongoDB Atlas can streamline the approach to gen AI, creating an effective and immediate positive impact to business in several ways: Reduced product onboarding time and costs: Retailers can onboard products faster and quickly make them available on their sales channels because of the ready-to-use tools and prebuilt integrations. This cuts down on the need for custom code and significantly lowers development costs. Increased flexibility and customization: MongoDB’s flexible document model and Together AI’s inference engine enables retailers to mold their applications to fit specific needs, such as back-office data processing, demand forecasting, and pricing as well as customer-facing conversational AI. Seamless integration with existing systems: MongoDB Atlas, in particular, integrates seamlessly with other frameworks we’re already using, like LangChain and LlamaIndex. This has made it easier to bring AI capabilities to adopt across various business units. Added support and expertise: The MongoDB AI Applications Program (MAAP) is especially helpful in beginning the journey into AI adoption across enterprises. It offers not just architectural guidance but also hands-on support, so enterprises can implement AI projects with confidence and a well-defined road map. Combining Together AI and MongoDB Atlas for a powerful approach to retail Together AI and MongoDB Atlas are a powerful combination for anyone in the retail industry looking to make the most of gen AI. It is evident how they help unlock valuable use cases, from personalized customer experiences to real-time operational improvements. By adopting MongoDB Atlas with Together AI, retailers can innovate, create richer customer interactions, and ultimately gain a competitive edge. If you’re exploring gen AI for retail, you’ll find that this combination has a quick, measurable, and transformative impact. Learn more about Together AI by visiting www.together.ai . For additional information, check out Together AI: Advancing the Frontier of AI With Open Source Embeddings, Inference, and MongoDB Atlas .

January 13, 2025
Next →

Leveraging an Operational Data Layer for Telco Success

The emergence of 5G network communication, IoT devices, edge computing, and AI have accelerated structural changes within the telecommunications industry, creating new needs and opportunities. To remain competitive, telcos must embrace this technology-driven transformation by defining a robust data strategy. Such a strategy should enhance operational efficiency and provide unique value to customers, and should ultimately enable telcos to set themselves apart from their competitors. All of this can be attained by leveraging an operational data layer (ODL) with MongoDB. Operating a consolidated ODL opens new business opportunities that telcos can incorporate into their value matrix, including customer support systems, AI-enriched applications, and IoT-oriented services. These unlocked capacities will help telecommunications companies succeed in a competitive market. Understanding the operational data layer An ODL is an architectural pattern that centrally integrates and organizes siloed enterprise data, making it available to consuming applications. It acts as an intermediary between data producers and consumers. This architecture pattern is illustrated below: Figure 1. ODL sample reference architecture, using MongoDB In this diagram, MongoDB Atlas acts as the ODL, centrally integrating siloed data from multiple sources, including CRM, HR, and billing. Initially, data is extracted to the ODL, transformed according to established requirements, and then loaded to the MongoDB database. By means of delta loads, the ODL is kept in sync over time. Consuming applications, both operational and analytical, access the ODL through an API layer, which delivers a common set of methods for users, and enforces security standards throughout the organization. Enhancing operational efficiency with MongoDB and the ODL At its core, implementing an ODL with MongoDB provides access to a rich document model and a data developer platform that boosts operational efficiency and unlocks the value of previously siloed enterprise data. The ODL attains this efficiency through a set of key capabilities inherent to MongoDB. The ODL benefits from the flexibility of the document model that adapts its schema to any application requirement while supporting multiple data structures. This polymorphic structure allows variations from document to document liberating applications from rigid schemas and supporting merging from non-identical entities. Telcos gain speed in development—which translates to better performance—when accessing data through an ODL, as they avoid costly join operations required by legacy applications. MongoDB provides a unique place for data storage that can be accessed in a single database operation decreasing end-user response times. Telcos can leverage MongoDB’s versatility to cast multiple workloads, store any data type, and to adopt a rich query language that executes complex operations. Subsequently, the ODL accepts sophisticated query pipelines capable of processing text, images, videos, geospatial data, facet search, analytical transformations, time series, and more. Horizontal and vertical scalability empowers telcos to receive large data volumes and high traffic loads essential for modern applications. This mechanism is achieved through sharding, a process that partitions and distributes data across multiple nodes, accommodating fluctuating workload demands and enhancing overall system performance. An ODL running in MongoDB Atlas benefits from a multi-cloud strategy that allows deployments across multiple cloud providers. This approach mitigates vendor lock-in risks, grants global coverage, and adapts to infrastructure requirements—ensuring that applications adhere to cost constraints, achieve performance benchmarks, and maintain regulatory compliance. MongoDB provides a robust security framework for storing and managing sensitive data due to its built-in tools—including encryption, authentication, authorization, network security, and auditing—thus protecting data against information breaches. It also complies with important international regulations for telcos like the General Data Protection Regulation (GDPR) and the Payment Card Industry Data Security Standard (PCI DSS). MongoDB provides a modern data platform designed to build, manage and scale applications in a unified developer experience. The developer platform fosters innovation allowing developers to access a variety of features to manage their ODL including Atlas Vector Search, Atlas Monitoring, and Atlas Triggers, among others. Refer to our official documentation to learn more about MongoDB Atlas . Using the ODL to gain a competitive advantage Fostering operational efficiency through an ODL is the initial step toward opening a new business that will eventually translate into a competitive advantage. Accordingly, telcos need to develop their own strategies and capitalize on the benefits from these unlocked opportunities, differentiating themselves in the industry. Well-known telcos have already leveraged this approach, creating successful business outcomes. They consolidate single-view instances , concentrating information from different business lines—such as mobile, fixed lines, broadband, and TV/entertainment—into MongoDB Atlas. This environment is well-suited for building personalized customer management solutions, overcoming challenges with siloed data environments. These telcos choose MongoDB because it offers a flexible data model that facilitates data aggregation and horizontal scaling, allowing them to efficiently leverage customer data to build customer-centric applications. Additionally, leading telcos are leveraging AI to enhance their operations, safeguard their business, and improve their services. One prominent use of AI is fraud detection and prevention . This is a critical area that, if poorly managed, can lead to negative consequences like financial losses, unmeasurable reputational damage, and unhinged security network risks. A consolidated ODL serves as a gateway for implementing fraud detection measures. Nowadays, MongoDB’s platform is ingesting and storing terabytes of data from multiple platforms to leverage AI models, potentially saving millions of dollars for telcos. Refer to our ebook, Innovate with AI: The Future Enterprise , to learn more. Telcos are also capitalizing on their networks and the MongoDB ODL by effectively managing the vast amounts of data generated by IoT devices, and adding new end-to-end services. MongoDB is helping large telcos effectively implement IoT platforms supplying scalability for growing device demand, flexibility to manage data model changes, and automatic data tiering to reduce storage costs. These capabilities ultimately improve customer experiences and speed time to market for new applications. Furthermore, ODLs improve product catalog management systems, which are increasingly common in the industry due to telcos’ expanding their offering to a broader set of products, from phone plans to bundled entertainment services. ODLs upgrade the product catalog, allowing for real-time product personalization and analytics. MongoDB assists telcos in upgrading their product catalog systems, enabling advanced search capabilities, reducing development time, and supporting seasonal workload demand. Refer to our white paper, Implementing an Operational Data Layer for Product Catalog Modernization , to learn more. Finally, an ODL accelerates the modernization of monolithic relational database systems that struggle to manage exponential data growth and to adapt to evolving business needs. Telcos use MongoDB in their modernization efforts to deliver 3 to 5x faster operations, allowing scaling to millions of records per day, while at the same time reducing their costs—typically by 50% or more. Future directions This blog highlights how implementing an ODL with MongoDB can unlock telcos’ ability to achieve operational efficiency through the native capabilities of MongoDB and its cloud offering. This innovative architecture not only improves operations, but also unlocks business opportunities that are the foundation for new competitive advantages. These enhanced capabilities represent the backbone to consolidate telcos’ strategic positioning, ultimately differentiating from their competitors in powerful ways. Visit our MongoDB for Telecommunications solutions page to learn more. If you would like to learn more about implementing an ODL with MongoDB for your TELCO organization, visit the following resources: White paper: Implementing an Operational Data Layer White paper: Unleash Telco Transformation with an Operational Data Layer Head over to our quick-start guide to get started with Legacy Modernization today.

January 14, 2025