Overview
The article discusses the engineering behind LinkedIn Connected, a new iOS app that utilizes anticipatory computing to enhance user networking experiences. It introduces the Relationship Opportunities On Demand (ROPOD) platform, which includes components like Synthesizer, SCRAM, and Opportunist to deliver timely notifications and relevant updates to users.
What You'll Learn
1
How to integrate anticipatory computing features into mobile applications
2
Why using a message broker like Kafka is essential for decoupling application components
3
How to implement a custom background process scheduler for mobile apps
Prerequisites & Requirements
- Understanding of anticipatory computing concepts
- Familiarity with Kafka and Rest.li
Key Questions Answered
What is the purpose of the LinkedIn Connected app?
LinkedIn Connected is designed to help users engage with their network effortlessly, providing timely notifications and relevant reasons to reach out. It replaces LinkedIn Contacts and enhances networking through anticipatory computing features.
How does the ROPOD platform function?
The Relationship Opportunities On Demand (ROPOD) platform consists of three components: Synthesizer, which aggregates opportunities; SCRAM, which identifies significant changes in user context; and Opportunist, which sends timely notifications. Together, they enhance user engagement and networking.
What technologies are used in the LinkedIn Connected app?
The app uses Rest.li for API communication and Kafka as a message broker to facilitate interaction between its components. This architecture allows for scalable and efficient processing of user data and notifications.
How does the Synthesizer component work?
The Synthesizer aggregates and ranks opportunities for users based on LinkedIn data. It utilizes a relevance model that combines both online and offline features, with daily workflows generating member-specific data stored in Voldemort, a key-value storage system.
Key Statistics & Figures
Percentage of LinkedIn members using mobile devices
over 50%
This statistic highlights the growing trend of mobile usage among LinkedIn members, emphasizing the need for mobile-optimized applications.
Technologies & Tools
Some links below are affiliate links. We may earn a commission if you make a purchase.
Backend
Rest.li
Used for API communication between components of the LinkedIn Connected app.
Backend
Kafka
Serves as a message broker to facilitate communication and decouple processing between different app components.
Database
Voldemort
A key-value storage system used for storing member-specific offline features.
Key Actionable Insights
1Implement anticipatory computing features to enhance user engagement in mobile applications.By anticipating user needs and providing timely information, apps can significantly improve user satisfaction and retention.
2Utilize Kafka as a message broker to decouple application components for better scalability.Decoupling allows each component to scale independently, which is crucial for handling varying loads and improving overall application performance.
3Design a custom background process scheduler to optimize power consumption in mobile apps.Intelligent scheduling of background tasks can enhance user experience by balancing performance with battery life.
Common Pitfalls
1
Failing to properly integrate anticipatory computing features can lead to irrelevant notifications.
Without a well-defined relevance model, users may receive notifications that do not align with their needs, leading to frustration and disengagement.
Related Concepts
Anticipatory Computing
Mobile Application Development
User Engagement Strategies