September 2024 Summaries

Ramblr, a San Francisco-based startup, is using its powerful and versatile 3D annotation and recognition capabilities to bring the benefits of gen AI into the physical world. The company's technology allows continuous recording of activities, with users asking computers in natural language about their performance or what they should do next. Ramblr has worked with international chemical giant BASF, using its software to assess footage from pipeline connection work and identify any errors. MongoDB Atlas is part of the Ramblr technology stack, providing flexible data storage options and built-in vector database capabilities. The company plans to expand into robotics and spatial computing sectors in the future.

An automotive giant has developed an AI-powered solution to revolutionize their diagnostic process for engine issues, aiming to increase efficiency, customer satisfaction, and revenue throughput. The project involved developing an AI-powered solution that could quickly analyze engine sounds and compare them to a database of known problems, significantly reducing diagnostic times. Despite initial setbacks, the project team, with the help of MongoDB's client in automotive manufacturing and PureInsights, managed to develop a groundbreaking solution that combines cutting-edge AI technologies with MongoDB's powerful data management capabilities. The result is a diagnostic tool leveraging text and audio analysis to significantly reduce diagnostic times, increase customer satisfaction, and boost revenue through the dealer network.

Anaiya Raisinghani discusses the appeal of "dirty" sodas, a unique beverage trend from Utah, popularized by Swig, which combines soda with cream and flavored syrups. Inspired by the Hulu show "The Secret Lives of Mormon Wives," she explores how to use technology to discover seasonal drink preferences from Swig's menu. The process involves scraping the website using Microsoft's Playwright to handle dynamic content, utilizing OpenAI's Structured Outputs to categorize drinks seasonally, and leveraging MongoDB Atlas Search to filter sodas based on ingredients and cravings. The tutorial provides a detailed guide on implementing these technologies to create a database of Swig's menu items, ultimately showcasing how to find the perfect drink for each season.

In this article, Apoorva Joshi, Senior AI Developer Advocate at MongoDB, shares her experiences with achieving a healthy work-life balance. She emphasizes the importance of setting boundaries and prioritizing activities outside of work to maintain productivity and overall well-being. Apoorva suggests incorporating everyday tasks into one's work calendar as a reminder to take breaks and communicate these boundaries with colleagues. By doing so, she has been able to create a fulfilling life that balances work, family, exercise, hobbies, and chores.

Pathfinder Labs, a New Zealand-based company specializing in software for law enforcement agencies and investigators, has developed Paradigm, a tool built on MongoDB Atlas that helps collect, gather, and convert data into actionable decisions. The platform is used by prestigious organizations globally, including Interpol. Paradigm simplifies evidence gathering and analysis, integrates external data sources and vendors, future-proofs investigation methods, and minimizes overall costs. It also incorporates AI capabilities such as MongoDB Vector Search to identify relevant information, classify documents and text, and build better search experiences for users. The team has found great value in MongoDB's Performance Adviser and Schema Anti Patterns features.

Glyphic AI, an AI-powered sales co-pilot founded by former engineers from Google DeepMind and Apple, aims to revolutionize the sales industry by streamlining workflows and enhancing customer relationships. By leveraging large language models (LLMs) and working with private and dynamic data, Glyphic AI helps teams analyze sales processes and calls, providing tailored strategies for different customer personas and company types. Built on MongoDB, AWS, and Anthropic, the platform automatically breaks down sales calls using established methodologies like MEDDIC and integrates Atlas Vector Search to efficiently index and query high-dimensional vectors. Glyphic AI's Global Search feature uses this technology to allow users to ask strategic questions and retrieve data from numerous sales calls. The company relies on advanced cloud solutions for efficient data management and innovation, using MongoDB Atlas for its flexibility, scalability, and integration with features like Atlas Vector Search. Glyphic AI's next step is refining LLMs specifically for the sales context to deliver more strategic insights and enhance their product offerings in the evolving generative AI market.

MongoDB has introduced its new Application Delivery Certification program to provide comprehensive training for developers and application delivery leads, equipping them with the knowledge and skills needed to design, develop, and deploy modern solutions at scale. The certification includes exclusive online learning, hands-on labs, and a proctored exam covering topics such as storage optimization, querying, aggregation, retrieval-augmented generation (RAG), and architecting with Atlas Vector Search. Piloted with partners Accenture and Capgemini, the program is self-service and flexible to accommodate learners' schedules. Completing this certification may also qualify partners for MongoDB AI Applications Program (MAAP), providing access to a broader network of customers needing help building and deploying modern applications enriched with AI technology.

Vietnam's e-commerce industry is booming, with online spending growing by 79% in the first quarter of 2024 and projected to reach $32 billion by 2025. Ahamove, Vietnam's largest on-demand delivery company, handles over 200,000 daily e-commerce, food, and warehouse deliveries nationwide. To meet the rising demand for deliveries and new services, Ahamove moved to MongoDB Atlas on Amazon Web Services (AWS) in 2019. The scalability of MongoDB is crucial during peak times like Christmas or Lunar New Year when order volume surges. Geospatial queries enabled by MongoDB help locate drivers, map them to restaurants for faster deliveries, and track orders without relying on third-party services. Ahamove's generative AI chatbot integrated with restaurants' Facebook pages makes it easier for customers to handle the entire order process in natural language. The company is also working on personalized dish recommendations based on consumer preferences and timely notifications of incoming orders for merchants.

Qura, an AI-powered legal search engine based in Stockholm, Sweden, aims to revolutionize the way lawyers and researchers search for information by leveraging the power of Language Models (LLMs). Trained on millions of pages of publicly available legal texts, Qura's custom models enable efficient text searches on large corpuses of data. Unlike other generative AI applications, Qura focuses on providing users with the best sources and information rather than summaries or answers to their questions. The company has partnered with MongoDB since its inception, utilizing the document database for flexibility and fast search capabilities. Qura's larger goal is to bring about a new generation of intelligent search across various industries.

The blog post discusses various search methods such as text, vector, hybrid, and AI-powered search. Text search is a traditional method that allows users to find specific information within a large set of data by entering keywords or phrases. It is suitable for queries requiring exact matches where the overarching meaning isn't critical. Vector search, on the other hand, helps solve the challenge of providing relevant results even when the user may not know what they're looking for. It converts any type of media or content into a vector using machine learning algorithms and then searches to find results similar to the target term. Semantic search focuses on meaning and prioritizes user intent by deciphering not just what users type but why they're searching, in order to provide more accurate and context-oriented search results. Hybrid search combines the strengths of text search with the advanced capabilities of vector search to deliver more accurate and relevant search results. It shines in scenarios where there's a need for both precision (where text search excels) and recall (where vector search excels), and where user queries can vary from simple to complex, including both keyword and natural language queries.

Version 5.0 of the Laravel MongoDB integration introduces significant updates that align MongoDB with Laravel's conventions, simplifying developers' work with features like ID aliasing, enhanced date handling, and streamlined model customization. The update includes the removal of MongoDB-specific service providers due to compatibility with Laravel's classes, and changes to connection and logging systems. This release is part of a broader focus on improving MongoDB's capabilities, including its integration with LangChain for AI applications, offering tools like LangGraph for multi-agent orchestration and Hybrid Search Retriever for advanced retrieval patterns. Additionally, MongoDB is undergoing a leadership transition with Dev Ittycheria stepping down as CEO, to be succeeded by Chirantan "CJ" Desai, who brings extensive growth-at-scale experience from previous executive roles. This transition aims to guide MongoDB through its next phase, leveraging its strengths in data-intensive applications and AI to capture upcoming innovation waves, while maintaining a seamless leadership transition and continuity in strategic direction.

MongoDB has been developing tooling to help developers create advanced AI applications using LangChain, a framework for performing multi-agent orchestration. The integration of LangGraph into MongoDB simplifies the creation of applications using large language models (LLMs), including AI agents that require memory to maintain context across multiple interactions. Additionally, MongoDB has added two new custom Retrievers to the langchain-mongodb Python package, making it easier than ever to use the full capabilities of MongoDB Vector Search with LangChain. Furthermore, MongoDB now supports the LangChain Indexing API for seamless loading and synchronization of documents from any source into MongoDB, leveraging LangChain's intelligent indexing features.

The article discusses the evolution of the AI landscape and how companies are seeking tailored solutions for their unique challenges. It highlights the importance of specialized AI frameworks, platforms, and tools in driving innovation. The author mentions MongoDB's integration with LangChain as an example of combining technical strengths to create more sophisticated AI applications. Additionally, five new AI partners that offer product integrations with MongoDB are introduced, including BuildShip, Inductor, Metabase, Shakudo, and VLM Run. The article emphasizes the need for adaptable solutions that can grow and change as developers' needs evolve.

Agmeta.AI, a startup focused on enhancing Customer Lifetime Value (CLTV), aims to improve customer service centers by providing businesses with actionable insights to combat churn and identify key customers for upselling opportunities. Founded by Samir Agarwal, the company leverages machine learning and generative AI, backed by MongoDB Atlas, to understand not only the content of calls but also their context. This enables Agmeta's software to detect customer dissatisfaction even when a conversation appears positive on the surface. By providing businesses with a more accurate Customer Satisfaction (CSAT) score and insights into customer behavior, Agmeta helps companies minimize churn and grow CLTV.

Atlas Stream Processing is a native stream processing solution integrated into MongoDB Atlas that offers an easy developer experience, cost savings, and performance advantages when using Apache Kafka in applications. It extends the MongoDB Query API to enable continuous data processing and supports advanced use cases not possible with the Kafka Connector. The decision between Atlas Stream Processing and the Kafka Connector depends on factors such as ease of use, performance, and cost. Atlas Stream Processing is a fully managed service that charges per Stream Processing Instance worker, offering potential cost savings compared to the Kafka Connector's task-based pricing model.

MongoDB has introduced new updates to Atlas Charts that enhance security, billing visibility, and customization features. The passcode-protected public dashboards feature provides an extra layer of security for shared insights. The updated billing dashboard allows users to view resource tags data and billing data from all linked organizations. Additionally, hyperlink customization options have been introduced for table charts in Atlas Charts. These updates aim to improve the overall data analysis experience for users.

The global energy landscape is undergoing significant transformation, with energy consumption increasing by 2.2% in 2023. Smart meters play a vital role in this evolution, offering accurate billing, personalized tariffs, enhanced grid management, and energy efficiency insights. However, the widespread adoption of smart meters also presents significant data-related challenges, including data interoperability, management, and privacy concerns. MongoDB's Time Series collections provide a robust solution for storing and analyzing smart meter data, while Cedalo's MQTT Platform offers advanced features to meet the demands of large enterprises. Together, these technologies offer an end-to-end solution for smart meter data management, enabling power distribution companies to analyze trends, implement real-time monitoring, and make informed decisions regarding their infrastructure.

In the first article of "Away From the Keyboard" series, Anaiya Raisinghani, a Developer Advocate at MongoDB, shares her day-to-day responsibilities and how she establishes a healthy work-life balance. She uses nonrefundable workout classes and dinner reservations to help her step away from work and has developed strategies to ensure that when she logs off for the day, she stays logged off. Anaiya emphasizes the importance of having hobbies and building relationships within one's community to be a happier human being. She advises developers seeking better balance to remove Slack from their personal devices or use separate devices for work and personal life.

MongoDB has announced a strategic partnership with Ditto, a company specializing in fast data synchronization for mobile and edge solutions, allowing enterprises to operate without reliance on Wi-Fi or cloud connections. Ditto's technology creates a distributed wireless network that facilitates peer-to-peer data synchronization, ensuring continued functionality even in low-connectivity environments. This collaboration promises to enhance mobile applications across various sectors, such as retail and banking, by enabling seamless service delivery and operations regardless of connectivity issues. Additionally, MongoDB's leadership transition is set to take place with Chirantan "CJ" Desai succeeding Dev Ittycheria as CEO, effective November 2025. CJ brings extensive experience from ServiceNow and Cloudflare, positioning MongoDB to continue its growth and innovation journey. Dev Ittycheria, who has been integral to MongoDB's development over the past eleven years, emphasizes the strategic timing of the leadership transition and expresses confidence in MongoDB's future under CJ's guidance.