Best of ByteByteGo — September 2025

1. 1
   Article

   ByteByteGo·32w

   How Grab Built An Authentication System for 180+ Million Users

   Grab transformed their fragmented authentication system serving 180+ million users by adopting OpenID Connect (OIDC) and implementing Dex, an open-source federated identity provider. The solution unified authentication across internal and external applications, replacing multiple custom OAuth implementations with a standardized approach. Key features include token delegation for service-to-service communication, multi-IdP failover for high availability, and single sign-on capabilities. This centralized system improved security, reduced administrative overhead, and provided a consistent user experience across Grab's ecosystem of ride-hailing, payments, and delivery services.

   133

   1
2. 2
   Article

   ByteByteGo·29w

   How Grab’s Migration from Go to Rust Cut Costs by 70%

   Grab successfully migrated their Counter Service from Go to Rust, achieving a 70% cost reduction through improved resource efficiency. While latency remained similar, the Rust version required only 4.5 CPU cores compared to Go's 20 cores for handling 1,000 requests per second. The migration involved careful library selection, rebuilding internal tools like configuration systems, and overcoming challenges with Rust's borrow checker and async concurrency model.

   94

   3
3. 3
   Article

   ByteByteGo·32w

   How Netflix Tudum Supports 20 Million Users With CQRS

   Netflix redesigned their Tudum platform architecture to support 20 million users by replacing a traditional CQRS implementation with RAW Hollow, an in-memory object store. The original design used Kafka and Cassandra with caching layers, causing delays in editorial previews due to eventual consistency. By embedding RAW Hollow directly into microservices, they eliminated external datastores and reduced page construction time from 1.4 seconds to 0.4 seconds while enabling near-instant editorial previews. The compressed in-memory approach stores three years of data in just 130MB while maintaining strong consistency options for critical workflows.

   87

   1
4. 4
   Article

   ByteByteGo·31w

   How Anthropic Built a Multi-Agent Research System

   Anthropic built a multi-agent research system using an orchestrator-worker pattern with a Lead Researcher agent coordinating specialized subagents and a Citation Agent for accuracy. The system outperformed single-agent setups by 90% through parallel processing and dynamic adaptation, though it consumes 15x more tokens. Key engineering principles include proper prompt design, delegation strategies, effort scaling, and parallelization. Production challenges involve managing stateful agents, debugging non-deterministic behavior, and handling deployments without breaking running tasks.

   74