Leveraging RAG-powered LLMs for analytical tasks

Summary

Data analysts at Grab faced an increasing volume of stakeholder data requests that required manually writing and executing repetitive SQL queries with minor parameter changes. To automate routine tasks like metric reporting and fraud investigations, the Integrity Analytics team paired an internal prompt-management platform, Spellvault, with Data-Arks, an in-house Python-based middleware that packages SQL queries and Python functions into APIs. A scheduler triggers automated report summarization, while an analytical bot named A* bot executes relevant investigation queries via RAG and returns summarized findings in Slack. Automated report generation saves an estimated three to four hours per report and shortens multi-query fraud investigations to minutes. Grab favored RAG over fine-tuning because it avoids continuous retraining costs, scales faster across diverse use cases, and surfaces current production data.

Context

Data analysts struggled with a growing number of stakeholder data requests, spending substantial time manually writing and executing similar SQL queries with minor parameter modifications for recurring reports and fraud investigations.

Approach / What changed

Grab integrated Spellvault, an internal prompt-management and low-code RAG platform, with Data-Arks, a Python API middleware that turns uploaded SQL queries and scripts into callable endpoints, orchestrating tasks via a scheduler and Slack bots.

Takeaways

• The automated Report Summarizer calling Data-Arks APIs and LLM summarization saves an estimated 3 to 4 hours per report.
• Data-Arks integrates with Slack, Wiki, and Jira APIs, enabling employees across teams to self-serve by turning custom SQL queries and Python scripts into callable APIs.
• Grab selected RAG over fine-tuning to eliminate expensive model retraining cycles and allow LLM applications to retrieve real-time data directly from production databases.

Related reading

• Grab · Nov 22, 2024

  How we reduced initialisation time of Product Configuration Management SDK

  GrabX operates as Grab's central platform for product configuration management, where client services fetch configuration data via an eventually consistent SDK. Services handling around 400 MB of configuration data experienced startup cold starts taking approximately four minutes, creating service stress during traffic spikes. The engineering team resolved this bottleneck through a multi-phase optimization of how the SDK retrieves data from AWS S3. First, sequential downloads of common and service-specific datasets were replaced with concurrent fetching. Next, concurrent downloading and memory loading were applied across large configurations within subscribed services, followed by the complete removal of an outdated disk-caching fallback mechanism. Benchmarks across diverse configuration payloads showed an overall initialisation time reduction of up to 90%.

  Ram Dilip Pradhan

• Grab · Feb 19, 2025

  Grab AI Gateway: Connecting Grabbers to multiple GenAI providers

  Grab built the AI Gateway to centralize access, cost control, and security across external and open-source Generative AI providers such as OpenAI, Azure, AWS, and Google. Designed as a set of lightweight reverse proxies, the gateway manages authentication, rate limiting, and authorization while translating payloads into a unified OpenAI-compatible interface. The platform archives request metadata and calculated per-call costs into a central data lake for auditing and showback, dynamically routing traffic across shared reserved capacity and regions to mitigate quota throttling. Supporting over 300 internal use cases, the system integrates directly with internal development notebooks and deployment tools to power applications ranging from real-time audio safety analysis to automated content moderation.

  Bjorn Jee

• Grab · Dec 26, 2024

  Embracing passwordless authentication with Grab’s Passkey

  Grab introduced Passkey to replace vulnerable traditional passwords and cumbersome multi-factor methods with a seamless, phishing-resistant alternative based on the FIDO standard. The architecture relies on an authenticator located on the user's device, a frontend client, and a backend storing only public keys and metadata. During registration and login, the frontend invokes WebAuthn APIs such as navigator.credentials.create and navigator.credentials.get using server-generated challenges to prevent replay attacks. Passkeys synchronize across ecosystems via Google Password Manager and Apple iCloud Keychain, allowing users to authorize logins with their device lock screen. This implementation improves user experience, eliminates the need to store secrets in backend databases, and cuts third-party communication costs associated with OTP delivery.

  Ocean Nguyen

• Grab · Sep 10, 2024

  Unveiling the process: The creation of our powerful campaign builder

  Grab details the event processing architecture behind Trident, its internal marketing campaign platform that evaluates If This, Then That (IFTTT) logic over Kafka streams. The core processing unit is a treatment consisting of an event, optional conditions, and actions. Complex campaign capabilities—such as counters, limits, and delays across multiple hours via recursive SQS message scheduling—are assembled from multiple coordinated treatments. To simplify campaign creation, Grab introduced a flowchart-like visual builder represented as a JSON node tree that compiles recursively into treatments while persisting node-to-treatment mappings to reconcile edits over time.

  Jie Zhang