Discover how Grab tackled the challenges of scaling its flagship membership program, GrabUnlimited. In this deep dive, we explore the migration from a legacy system to Temporal, reducing production…

In the world of high-volume data processing, migrating services without disruption is a formidable challenge. At Grab, we recently undertook this task by splitting one of our backend service's stream…

Discover how Grab's LLM-Kit enhances AI app development by addressing scalability, security, and integration challenges. This article discusses the challenges faced in LLM app building, the solution,…

Discover how we revolutionised our product configuration management SDK, reducing initialisation time by up to 90%. Learn about the challenges we faced with cold starts and the phased approach we took…

In the initial article, we explored the integration of Large Language Models (LLM) to automate metadata generation, addressing challenges like limited customisation and resource constraints. This…

Learn about GrabX, Grab’s central platform for product configuration management. This article discusses the steps taken to optimise the SDK, aiming to improve resource utilisation, reduce costs, and…

Vector similarity search has revolutionised data retrieval, particularly in the context of Retrieval-Augmented Generation in conjunction with advanced Large Language Models (LLMs). However, it…

Retrieval-Augmented Generation (RAG) is a powerful process that is designed to integrate direct function calling to answer queries more efficiently by retrieving relevant information from a broad database. In the rapidly evolving business landscape, Data Analysts (DAs) are struggling with the growing number of data queries from stakeholders. The conventional method of manually writing and running similar queries repeatedly is time-consuming and inefficient. This is where RAG-powered Large Language Models (LLMs) step in, offering a transformative solution to streamline the analytics process and empower DAs to focus on higher value tasks.

In this article, we will share how the Integrity Analytics team has built out a data solution using LLMs to help automate tedious analytical tasks like generating regular metric reports and performing fraud investigations.

As Southeast Asia’s leading super app, Grab serves millions of users across multiple countries every day. Our services range from ride-hailing and food delivery to digital payments and much more. The backbone of our operations? Machine Learning (ML) models. They power our real-time decision-making capabilities, enabling us to provide a seamless and personalised experience to our users. Whether it’s determining the most efficient route for a ride, suggesting a food outlet based on a user’s preference, or detecting fraudulent transactions, ML models are at the forefront.

However, serving these ML models at Grab’s scale is no small feat. It requires a robust, efficient, and scalable model serving platform, which is where our ML model serving platform, Catwalk, comes in.

Catwalk has evolved over time, adapting to the growing needs of our business and the ever-changing tech landscape. It has been a journey of continuous learning and improvement, with each step bringing new challenges and opportunities.

In May 2023, Grab unveiled the Live Activity feature for iOS, which received positive feedback from users. Live Activity is a feature that enhances user experience by displaying a user interface (UI) outside of the app, delivering real-time updates and interactive content. At Grab, we leverage this feature to keep users informed about their order updates without requiring them to manually open the app.

While Live Activity is a native iOS feature provided by Apple, there is currently no official Android equivalent. However, we are determined to bring this immersive experience to Android users. Inspired by the success of Live Activity on iOS, we have embarked on design explorations and feasibility studies to ensure the seamless integration of Live Activity into the Android platform. Our ultimate goal is to provide Android users with the same level of convenience and real-time updates, elevating their Grab experience.

In our system, we use the term “treatment” to refer to the core unit of a full IFTTT data structure. A treatment is an amalgamation of three key elements - an event, conditions (which are optional), and actions. For example, consider a promotional campaign that offers “100 GrabPoints for completing a ride paid with GrabPay Credit”. This campaign can be transformed into a treatment in which the event is “ride completion”, the condition is “payment made using GrabPay Credit”, and the action is “awarding 100 GrabPoints”.

Data generated across various Kafka streams by multiple services within Grab forms the crux of events and conditions for a treatment. Trident processes these Kafka streams, treating each data object as an event for the treatments. It evaluates the set conditions against the data received from these events. If all conditions are met, Trident then executes the actions.

Key to innovation and improvement in machine learning (ML) models is the ability for rapid iteration. Our team, Chimera, part of the Artificial Intelligence (AI) Platform team, provides the essential compute infrastructure, ML pipeline components, and backend services. This support enables our ML engineers, data scientists, and data analysts to efficiently experiment and develop ML solutions at scale.

With a commitment to leveraging the latest Generative AI (GenAI) technologies, Grab is enhancing productivity tools for all Grabbers. Our Chimera Sandbox, a scalable Notebook platform, facilitates swift experimentation and development of ML solutions, offering deep integration with our AI Gateway. This enables easy access to various Large Language Models (LLMs) (both proprietary and open source), ensuring scalability, compliance, and access control are managed seamlessly.

Dive into our journey of improving in-app translation experience amidst a post-COVID tourism boom. Discover how we overcame language detection hurdles, crafted an in-house translation model, and…

通过启用PGO（Profile-guided Optimization），可以显著减少CPU和内存的使用。在实验中，启用PGO后，CPU使用率降低了至少10%，内存使用量减少了至少10GB（30%）。此外，通过实例TalariaDB的应用，启用PGO后，存储事件的卷使用量减少了至少7GB（38%）。然而，在Catwalk服务中，启用PGO所需的工作量可能不值得获得的改进。总的来说，启用PGO的适用性和效益因服务的特性、当前架构和支持机制而异。未来，随着更多服务对PGO的支持和改进，可能会实现更广泛的PGO应用，提供更快的响应时间、更低的资源消耗和更好的用户体验。

Discover how Grab assesses marketing effectiveness using advanced attribution models and strategic testing to improve campaign precision and impact.

Join us as we share our experience in developing and implementing a consistent upgrade routine. This process underscored the significance of adaptability, comprehensive preparation, efficient…