Articles from Source: Lyft-Engineering

🚀 At Lyft, data is central to operations. To address inconsistencies in metric definitions across teams, we developed the Metric Semantic Layer (MSL). 📊 MSL serves as a centralized repository for all metric definitions, ensuring clear communication and decision-making. Key principles include: 1. Simplified onboarding and change management. 2. Intentional governance for data quality. 3. Transparency and accessibility for all users. 🔍 This system enhances collaboration and consistency in data...

🚀 At Lyft, we transformed our chaotic support ticketing system into a streamlined, self-routing operation over five years. Initially, our Jira Help Center was cumbersome with duplicate forms and no clear routing logic. We consolidated four portals into one dynamic form, enhancing efficiency and visibility. Now, one-third of tickets are routed automatically, saving hours of manual work annually. This evolution not only improved response times but also empowered our teams to focus on analytics....

🚗✨ Lyft has improved the pickup experience for riders in gated communities, addressing common frustrations. Previously, drivers often couldn't access gated areas, leading to cancellations and stress. The Mapping team has implemented key changes, including smarter pickup spot selections and better communication of gate access instructions. Now, riders can choose to meet drivers outside the gate, enhancing convenience. This initiative aims to minimize delays and improve overall satisfaction for...

At Lyft, predicting rider conversion is crucial for optimizing user experience and balancing supply and demand. 🚗 The challenge lies in data sparsity; specific contexts often provide limited data, making accurate predictions complex. To address this, Lyft developed a Bayesian Tree model, which structures data hierarchically and uses Bayesian smoothing for better predictions even in sparse situations. 📊 This innovative approach enhances real-time decision-making while ensuring predictions...

🚗 Lyft employs a complex system to balance rider demand and driver supply through pricing and incentives. Understanding the long-term effects of these decisions is critical. The Foundational Models team uses a two-step approach to measure "market-mediated long-term effects" based on user experiences. This involves estimating how policy changes impact negative experiences and how these experiences influence future behavior. Their methodology allows for continuous calibration of decisions,...

🚀 Lyft has revamped its localization process to meet growing demands for speed and quality. By integrating AI with human oversight, Lyft's new translation pipeline reduces turnaround times from days to minutes. This change supports its expansion into new markets and compliance with local regulations. The pipeline operates in three phases: drafting, early release, and final review, ensuring high-quality translations. Learn more about this innovative approach! 🌍💡 #Localization #AI #Translation...

🚗📊 Lyft explores the use of quasi-experimental methods like Augmented Inverse Propensity Weighting (AIPW) to measure causal impacts when A/B testing isn't feasible. These methods help assess partnerships, long-term effects, and data biases. Validation and diagnostics are crucial to ensure accurate results, focusing on confounders and model integrity. Learn more about the importance of trust in non-randomized measurements! #CausalInference #DataScience #QuasiExperiments #LyftEngineering

🚗 Lyft's Feature Store is a key element of its Data Platform, designed to streamline Machine Learning (ML) feature management at scale. This system centralizes feature engineering, ensuring consistency across diverse models and facilitating efficient model training and inference. The architecture includes Batch, Online, and Streaming features, enhancing user experience and accessibility for engineers. For more insights on the evolution and impact of the Feature Store, check out the full...

🚀 Upgrading from Python 3.8 to 3.10 revealed a memory leak at Lyft. During the upgrade, we noticed increased latency and timeouts in one service, linked to repository queries causing thread join delays. Using our internal memory profiling tool, we traced the issue to a compatibility problem between gevent and urllib3. Downgrading urllib3 resolved the leak! For memory leak issues, consider using gunicorn's max-request settings to prevent OOM errors. #Python #MemoryManagement #LyftEngineering...

🚀 At Lyft, machine learning drives key operations like dispatch and pricing. As our platform expanded, we faced challenges with complexity and scalability. 📈 To tackle this, we restructured LyftLearn from a fully Kubernetes-based system to a hybrid model. This combines AWS SageMaker for offline tasks and Kubernetes for online serving, optimizing performance and simplifying architecture. 🔍 The transition involved significant technical adjustments, ensuring seamless workflows for our data...

🚗 Excited to share insights from my journey as a Data Scientist at Lyft, focusing on my starter project with the Rider Experience Score (RES) tool. RES helps measure the long-term impact of rider experiences, like ETA, on retention without relying solely on A/B tests. 📊 I navigated various challenges to improve the RES pipeline, collaborating with colleagues to identify key rider experiences and ensure accurate causal estimates. Lyft is hiring! If you're interested in Data Science, check out...

🚀 Lyft has successfully migrated its Android codebase to Kotlin, a journey that began in 2018. The Rider, Driver, and Urban Solutions apps are now fully Kotlin-based. This transition offers benefits like concise code, faster compile speeds with the K2 compiler, and support for modern UI frameworks like Compose. To manage the migration, Lyft utilized a tool called Migration Tracker, which monitors progress and helps automate the process. Challenges included issues with the migration tool and...

🚗✨ Discover the journey of two data scientists, Morteza Taiebat and Han Gong, who started as interns at Lyft and now contribute to impactful projects. They share insights on their experiences, focusing on sustainability and driver loyalty. Their work includes evaluating electric vehicle (EV) adoption and designing incentive strategies for drivers. If you're considering an internship at Lyft, their stories highlight the growth opportunities and collaborative culture that await you! 🌱📊...

🚗💡 Ridesharing platforms like Lyft tackle complex matching challenges daily. Each rider and driver represents a unique piece in a dynamic puzzle, requiring real-time solutions for efficient urban mobility. Graph theory helps model these matches, particularly through bipartite graphs. This allows for flexible connections based on factors like distance and time. Lyft's dispatch team continually processes millions of potential decisions, aiming to optimize pickups and driver earnings. Stay tuned...