How Pinterest Uses Thrift

This article discusses the re-architecture of the serving stack for next-generation ads lightweight ranking models at Pinterest, moving from a traditional Two-Tower architecture to a more complex G...

The article reflects on a decade of AI platform development at Pinterest, detailing the evolution from fragmented machine learning stacks to a unified AI platform that supports various models.

Pinterest successfully migrated 3. 7 million lines of code from Flow to TypeScript over eight months, enhancing type safety and developer experience.

The article discusses the Structured DataStore (SDS), a unified multi-model data management platform developed by Pinterest.

The article discusses Pinterest's implementation of feature caching in their recommender systems using Cachelib, an in-process caching engine developed by Meta Open Source.

This article discusses the redesign of Pinterest's ad serving systems, focusing on achieving zero downtime during the transition.

The article discusses Pinterest's adoption of TiDB as a replacement for HBase, detailing the motivations, selection methodology, and the journey of integrating TiDB into their infrastructure.

The article discusses the redesign of the Goku-Ingestor at Pinterest, focusing on enhancing performance, reducing costs, and minimizing code complexity.

This article discusses the integration of Python-based anomaly detection algorithms into Pinterest's Warden platform, originally built in Java.

This article discusses the debugging process of a direct memory leak encountered in Apache Flink applications at Pinterest.

The article discusses improvements made to the training foundations of Pinterest's Closeup Recommendation Ranker, focusing on data logging, sampling strategies, and an auto-retraining framework.

The article discusses Pacer, Pinterest's new asynchronous computing platform designed to address the limitations of its predecessor, Pinlater.

The article discusses the development of a large-scale user signal platform at Pinterest, which enables real-time indexing of user events and construction of user sequences for machine learning app...

This article discusses the process of migrating data from HBase to TiDB at Pinterest with zero downtime.

This article is the second part of a series focused on debugging deadlock issues in the PininfoService during an upgrade to Ubuntu 18.

This article discusses Pinterest's transition to using Druid as its next-generation analytics data store, detailing the architecture and optimization strategies for host types.

This article discusses Pinterest's transition to using Druid as their analytical data store, detailing the challenges faced with HBase, the architecture of their Druid implementation, and insights ...

The article discusses how Pinterest utilizes Apache Spark SQL for interactive querying, detailing the architecture, challenges faced, and solutions implemented to enhance user experience.

The article discusses how Pinterest improved data processing efficiency by implementing partial deserialization of Thrift encoded data.

The article discusses Pinterest's efforts to develop fast and reliable iOS builds by optimizing their build process and adopting Bazel as their primary build tool.

This article discusses the development of a label-based enforcement pipeline at Pinterest aimed at enhancing Trust & Safety.

The article discusses Pinterest's transition from Lambda architecture to Kappa architecture for visual signals infrastructure, focusing on the need for real-time processing of machine learning sign...

The article discusses the importance of upgrading outbound Pin links from HTTP to HTTPS to enhance user security on Pinterest.

PinalyticsDB is Pinterest's proprietary time series database built on HBase, designed to store and visualize a large volume of time series reports.

The article discusses Pinterest's implementation of Presto, an open-source distributed SQL query engine, detailing the challenges faced and solutions developed to manage large-scale data analysis.

The article discusses the open-sourcing of Singer, Pinterest's logging agent designed for efficient and reliable data collection from numerous hosts.

The article discusses how Pinterest improved the speed and reliability of its iOS builds by adopting Bazel as their build system.

The article discusses the challenges Pinterest faced with their Apache Thrift schemas, which had become tightly coupled and complex, leading to inefficiencies in development cycles.

The article discusses the development and implementation of the Linchpin Domain-Specific Language (DSL) at Pinterest, which streamlines the process of building and deploying machine learning models...

The article discusses Manas, a high-performing customized search system developed by Pinterest to handle billions of queries and improve search relevance among over 100 billion Pins.

The article discusses Pinterest's upgrade of HBase from version 0. 94. 26 to 1. 2, emphasizing the importance of maintaining high performance and availability during the transition.

The article discusses the applications of Pintrace data at Pinterest, highlighting its role in improving backend service latency and debugging.

The article discusses enhancements made to Pinterest's security measures, specifically the introduction of email notifications for new logins and the ability for users to view and revoke active ses...

The article discusses Pinterest's implementation of distributed tracing using their open-source tool, Pintrace, which tracks requests across microservices in their backend.

The article discusses the open-sourcing of KingPin, a toolset developed by Pinterest to enhance scalability and reliability in their infrastructure.

The article introduces new open-source tools for the Elixir community, highlighting the advantages of Elixir and Erlang in handling concurrency and errors.

The article discusses the open-sourcing of PinLater, an asynchronous job execution system developed by Pinterest.

The article discusses the development and implementation of Aragog, a system designed by Pinterest to efficiently fetch, store, process, and serve billions of URLs at low latencies.

Skyline is an ETL-as-a-Service platform developed by Pinterest to streamline data processing and reporting for its users.

The article discusses the development of Pinalytics, Pinterest's customizable big data analytics platform designed to enhance data accessibility and analysis for employees.

The article discusses the rebuilding of Pinterest's user typeahead feature to enhance user experience in discovering contacts.