Highlights From MongoDB World 2022, Day 2

Steve Jurczak
June 9, 2022 | Updated: January 5, 2023
#MongoDB World

Day Two of MongoDB World 2022 was all about the breakout sessions — more than 80 were on tap for the day.

Things kicked off shortly after 8 a.m. with a discussion on empowering women and other underrepresented groups in the workplace, held in the IDEA Lounge. The 9 a.m. slot was packed with 10 sessions that ranged from building a sustainable ecosystem to the principles of data modeling to using Rust to build applications.

Steve Westgarth, senior director of engineering at GSK (formerly GlaxoSmithKline) dove into the weighty topic of morality in the digital world and what developers ought to do when the software they build leads to unintended consequences. All too often, there’s immense pressure to release MVPs early — before all potential vulnerabilities have been vetted. Westgarth’s session sprang from a rhetorical question: “Do we as engineers have an ethical and moral responsibility to anticipate unintended consequences and how much personal responsibility should an individual take to ensure ethical management of data?” His discussion answered that with a Yes — developers do have to weigh the risk of unintended consequences, such as data breaches, versus the desire to maximize market opportunity. Westgarth urged developers to ask themselves what the unintended consequences are of the software they have in production, and to raise awareness of these issues in their organizations.

A 15-minute lightning talk followed, with a session name that made it a popular draw for fans of worst-case scenarios: “Strange Cases From the Field.” Adam Schwartz, MongoDB director of technical services in EMEA, walked attendees through some especially challenging real-life technical support stories. He gave a detailed account of such curious cases as The Mistaken Hypotheses and The Unsuccessful Mitigations, and shared lessons he learned during years in the trenches as a support specialist. Closing on a positive note, he assured attendees that problem cases are rare, most cases have straightforward solutions, and exceptional cases are always a learning experience.

Day One saw Mark Porter announce the MongoDB Relational Migrator, including a live demo of the product. On Day Two, lead product manager Tom Hollander did a deep dive into use cases, justifications, and future capabilities for the tool. MongoDB Relational Migrator imports and analyzes relational database schemas, maps them to an appropriate MongoDB schema, and transforms and migrates the data into MongoDB. Hollander said organizations can experience a 3x to 5x increase in development velocity and up to 70% in cost reductions by migrating away from relational models in favor of a more modern deployment such as MongoDB Atlas. Hollander said he anticipates future capabilities to include continuous replication, Kafka integration, application code generation, schema recommendations, and more.

One company thriving in its legacy modernization efforts is Vodafone. The global head of engineering and transformation, Felipe Canedo, described Vodafone’s transition from a traditional telecommunications company to a Telco-as-a-Service (TaaS) provider. At the core of this transition was the creation of a scalable and open platform for the company’s engineers to innovate with complete freedom and flexibility. Canedo said Vodafone chose MongoDB because of its security, cloud-native high availability, support for multi-region and multi-cloud deployments, agile delivery, professional services, and ease of integration. The ultimate goal, Canedo said, was to provide Vodafone engineers with the best software experience possible.

Day One also saw MongoDB CPO Sahir Azam announce the general availability of MongoDB Atlas serverless instances. On Day Two, MongoDB advisory solutions architect Carlos Castro gave a live demo of deploying a serverless database. In 15 minutes, starting from the Atlas dashboard, Castro took the audience step-by-step through the process of selecting a cloud provider, spinning up the instance, creating an app service, authentication, and users, and then setting up rules to allow users to access data on the instance. Serverless instances always run the latest version of Atlas, include always-on security, and enable customers to only pay for operations they run.

Day Two also featured several discussions with leading experts and MongoDB partners. MongoDB senior vice president, product management, Andrew Davidson hosted a panel with three leaders in the effort to close the Developer Experience Gap: Peggy Rayzis, senior director of developer experience for Apollo GraphQL; Lee Robinson, director of developer relations for Vercel; and Søren Bramer Schmidt, chief architect and founder for Prisma.

Rayzis cited Apollo’s supergraph as one way it's helping developers be more productive by unlocking their flow state. “When you’re in that flow state, you’re writing better code, making better decisions, and developing better value for consumers,” she said.

Schmidt pointed out how the newest generation of developers stand to benefit the most from the proliferation of developer tools. “New generations of developers are much bigger and we can invest in better tooling for them,” Schmidt said. “It’s an exciting time to be building tools for developers.”

Lee emphasized the important role the open source community plays in these tools. “People hear about Vercel through Next.js,” Lee said, “and we invest to give back to the open source community.”

As gratifying and fun the first two days of World were, we really have something special in store for Day Three. It kicks off with a final keynote address by best-selling author, pioneering inventor, and futurist Ray Kurzweil.

Day Three also features our Builder’s Fest, where even MongoDB CTO Mark Porter is expected to lend his considerable expertise to a few promising projects. With live game shows, chaos presentations, nerd battles and more, MongoDB World 2022 will finish on a high note.

Check back tomorrow for more highlights from MongoDB World 2022.

← Previous

Highlights from MongoDB World 2022, Day 1

MongoDB World is back in person at New York’s Jacob Javits Center after a three-year hiatus. Day One featured a jam-packed schedule of educational sessions, live tutorials, customer stories, and product announcements for a crowd of nearly 2,700 developers and IT professionals. The developer-focused conference got off to an early start with breakout sessions beginning at 8 a.m. Three sessions were on tap: an introduction to data modeling with MongoDB, a primer on MongoDB Atlas Search , and a tutorial on getting started with MongoDB Atlas . In that tutorial, MongoDB solution architect Tom Gleitsmann explained how, out of all the challenges developers face on a daily basis, the common denominator is friction. Gleitsmann gave a crisp and informative summary of MongoDB Atlas features that were engineered specifically to reduce the amount of friction developers face, including ease of deployment, security by default, data visualization, the Performance Advisor , alerts, and backup scheduling, to name a few. The early-morning sessions were followed by a keynote delivered by MongoDB CEO Dev Ittycheria and Chief Product Officer Sahir Azam celebrating the company’s rapid growth, setting out a vision for its future, and highlighting several of its customers. The executives were joined on stage by Vercel founder and CEO Guillermo Rauch, Wells Fargo head of digital enablement Catherine Li, Avalara VP of software engineering John Jemseck, and several MongoDB product experts, each providing insight into the latest enhancements to MongoDB. The biggest reveal, though, was a new vision for MongoDB Atlas and the products that work seamlessly with it, such as Atlas Search and Atlas Data Federation . “We believe that developers want to build on a modern data model that's designed to the way they think and the way they code,” Ittycheria said. “And we also believe that developers want an elegant developer experience that makes their lives so much easier. And they want all this in one unified platform. What they need is a developer data platform.” After the morning keynote, sessions ran back-to-back until lunch. They ranged from quick, 15-minute “chalk talks” to hour-plus deep dives. In one, MongoDB software engineer James Wang gave a hands-on tutorial on using our data visualization tool, MongoDB Atlas Charts , which is fully integrated with MongoDB Atlas. Wang showed how easy it is to link data sources in just a few clicks. Using a fictitious company, he demonstrated step-by-step how to embed data visualization via code snippets and an SDK, share the data with others using a public link, filter data inside the admin web page, and restrict access to authorized users. Attendees followed along on their own laptops and were quickly able to replicate the visualizations. In another talk, Keller Williams’ senior architect Jim McClarty shared some of the real-world impact of Atlas — how it has accelerated the real estate firm’s ability to innovate its applications, how essential Atlas Search is in their applications, and how Charts has become “the best hidden feature in Atlas.” Attendees shuttled from room to room like they had places to go and people to meet, which they did. MongoDB principal, industry solutions, Felix Reichenback took attendees through mobile sync and why developers often waste tons of time trying to build their own sync tool that fails to handle conflict resolution because of the intermittent nature of mobile connections. Next, Michael van der Haven, VP at consulting giant CGI and expert in cloud-native platforms, explained how he helped the energy industry’s open source architecture group, OSDU, migrate away from Elasticsearch, simplify its architecture by removing memory-intensive indexes, and reduce OPEX by six figures using MongoDB Atlas. After lunch, MongoDB CTO Mark Porter gave an energetic keynote, announcing several more new products and features, including the new MongoDB Atlas CLI , the general availability of the Data API , and, perhaps our biggest announcement of the day, Queryable Encryption , which allows users to search their databases while sensitive data stays encrypted. Available in preview, Queryable Encryption offers a big step forward in protecting sensitive data. Porter gave personal anecdotes illustrating many of the hurdles developers have to overcome that have nothing to do with building software, such as rigid and fragile relational databases, and working with SQL, a language that developers early in their careers or fresh out of school have no desire to work with. Porter’s keynote address included a live demo of the Relational Migrator, which, while risky to perform in front of an audience, went off flawlessly. Meanwhile, a series of events kept the IDEA Lounge a lively place, including a great panel discussion called Our Journey: Being Black in Tech. And a floor below the workshops, more than a dozen MongoDB partners demonstrated their platforms and related products — including many of the companies named MongoDB Partners of the Year . The schedule for Day 2 is equally packed, with more than 80 sessions that include partner showcases, strange cases from the field, book club sessions, more deep dives into product announcements and tutorials, and talks on diversity, equity, and inclusion. In the afternoon, MongoDB celebrates Pride with food, drinks, and entertainment at the historic Stonewall Inn. And MongoDB World 2022’s biggest event happens at the end of the day — “The Party,” featuring music from The Midnight and Don Diablo, as well as retro arcade games and an open bar. Check back tomorrow for more highlights from MongoDB World 2022.

June 8, 2022
Next →

Retrieval Augmented Generation for Claim Processing: Combining MongoDB Atlas Vector Search and Large Language Models

Following up on our previous blog, AI, Vectors, and the Future of Claims Processing: Why Insurance Needs to Understand The Power of Vector Databases , we’ll pick up the conversation right where we left it. We discussed extensively how Atlas Vector Search can benefit the claim process in insurance and briefly covered Retrieval Augmented Generation (RAG) and Large Language Models (LLMs). MongoDB.local NYC Join us in person on May 2, 2024 for our keynote address, announcements, and technical sessions to help you build and deploy mission-critical applications at scale. Use Code Web50 for 50% off your ticket! Learn More One of the biggest challenges for claim adjusters is pulling and aggregating information from disparate systems and diverse data formats. PDFs of policy guidelines might be stored in a content-sharing platform, customer information locked in a legacy CRM, and claim-related pictures and voice reports in yet another tool. All of this data is not just fragmented across siloed sources and hard to find but also in formats that have been historically nearly impossible to index with traditional methods. Over the years, insurance companies have accumulated terabytes of unstructured data in their data stores but have failed to capitalize on the possibility of accessing and leveraging it to uncover business insights, deliver better customer experiences, and streamline operations. Some of our customers even admit they’re not fully aware of all the data in their archives. There’s a tremendous opportunity to leverage this unstructured data to benefit the insurer and its customers. Our image search post covered part of the solution to these challenges, opening the door to working more easily with unstructured data. RAG takes it a step further, integrating Atlas Vector Search and LLMs, thus allowing insurers to go beyond the limitations of baseline foundational models, making them context-aware by feeding them proprietary data. Figure 1 shows how the interaction works in practice: through a chat prompt, we can ask questions to the system, and the LLM returns answers to the user and shows what references it used to retrieve the information contained in the response. Great! We’ve got a nice UI, but how can we build an RAG application? Let’s open the hood and see what’s in it! Figure 1: UI of the claim adjuster RAG-powered chatbot Architecture and flow Before we start building our application, we need to ensure that our data is easily accessible and in one secure place. Operational Data Layers (ODLs) are the recommended pattern for wrangling data to create single views. This post walks the reader through the process of modernizing insurance data models with Relational Migrator, helping insurers migrate off legacy systems to create ODLs. Once the data is organized in our MongoDB collections and ready to be consumed, we can start architecting our solution. Building upon the schema developed in the image search post , we augment our documents by adding a few fields that will allow adjusters to ask more complex questions about the data and solve harder business challenges, such as resolving a claim in a fraction of the time with increased accuracy. Figure 2 shows the resulting document with two highlighted fields, “claimDescription” and its vector representation, “claimDescriptionEmbedding” . We can now create a Vector Search index on this array, a key step to facilitate retrieving the information fed to the LLM. Figure 2: document schema of the claim collection, the highlighted fields are used to retrieve the data that will be passed as context to the LLM Having prepared our data, building the RAG interaction is straightforward; refer to this GitHub repository for the implementation details. Here, we’ll just discuss the high-level architecture and the data flow, as shown in Figure 3 below: The user enters the prompt, a question in natural language. The prompt is vectorized and sent to Atlas Vector Search; similar documents are retrieved. The prompt and the retrieved documents are passed to the LLM as context. The LLM produces an answer to the user (in natural language), considering the context and the prompt. Figure 3: RAG architecture and interaction flow It is important to note how the semantics of the question are preserved throughout the different steps. The reference to “adverse weather” related accidents in the prompt is captured and passed to Atlas Vector Search, which surfaces claim documents whose claim description relates to similar concepts (e.g., rain) without needing to mention them explicitly. Finally, the LLM consumes the relevant documents to produce a context-aware question referencing rain, hail, and fire, as we’d expect based on the user's initial question. So what? To sum it all up, what’s the benefit of combining Atlas Vector Search and LLMs in a Claim Processing RAG application? Speed and accuracy: Having the data centrally organized and ready to be consumed by LLMs, adjusters can find all the necessary information in a fraction of the time. Flexibility: LLMs can answer a wide spectrum of questions, meaning applications require less upfront system design. There is no need to build custom APIs for each piece of information you’re trying to retrieve; just ask the LLM to do it for you. Natural interaction: Applications can be interrogated in plain English without programming skills or system training. Data accessibility: Insurers can finally leverage and explore unstructured data that was previously hard to access. Not just claim processing The same data model and architecture can serve additional personas and use cases within the organization: Customer Service: Operators can quickly pull customer data and answer complex questions without navigating different systems. For example, “Summarize this customer's past interactions,” “What coverages does this customer have?” or “What coverages can I recommend to this customer?” Customer self-service: Simplify your members’ experience by enabling them to ask questions themselves. For example, “My apartment is flooded. Am I covered?” or “How long do windshield repairs take on average?” Underwriting: Underwriters can quickly aggregate and summarize information, providing quotes in a fraction of the time. For example, “Summarize this customer claim history.” “I Am renewing a customer policy. What are the customer's current coverages? Pull everything related to the policy entity/customer. I need to get baseline info. Find relevant underwriting guidelines.” If you would like to discover more about Converged AI and Application Data Stores with MongoDB, take a look at the following resources: RAG for claim processing GitHub repository From Relational Databases to AI: An Insurance Data Modernization Journey Modernize your insurance data models with MongoDB and Relational Migrator

April 18, 2024