Skip to footer

Inside Uber ATG’s Data Mining Operation: Identifying Real Road Scenarios at Scale for Machine Learning

Uber ATG's self-driving vehicles measure a multitude of possible scenario variations to answer the age-old question: "how does the pedestrian cross the road?"

Meta-Graph: Few-Shot Link Prediction Using Meta-Learning

Uber AI introduces Meta-Graph, a new few-shot link prediction framework that facilitates the more accurate training of ML models that quickly adapt to new graph data.

Profiles in Coding: Christabelle Bosson, Uber Elevate, San Francisco

Christabelle Bosson, a senior advanced airspaces services engineer, discusses her journey from NASA to Uber Elevate, what excites her about the future of aerial ridesharing, and advice for aspiring aerospace engineers.

Developing the Next Generation of Coders with the Dev/Mission Uber Coding Fellowship

Participants in the Dev/Mission <> Uber Coding Fellowship took weekly courses taught by Uber engineers and worked with volunteers from Code for San Francisco on projects that benefit the local community.

Introducing Athenadriver: An Open Source Amazon Athena Database Driver for Go

Uber ATG built Athenadriver, an open source Amazon Athena database driver for Go, to facilitate communication between our business intelligence tools and the cloud.

Building Uber’s Go Monorepo with Bazel

Go and Bazel logos
When Uber adopted the open source Bazel build system, our engineers found many opportunities to contribute improvements to how Bazel works with a large Go monorepo.

Announcing a New Framework for Designing Optimal Experiments with Pyro

Uber AI released a new framework on top of Pyro that lets experimenters seamlessly automate optimal experimental design (OED) for quicker model iteration.

Monitoring Data Quality at Scale with Statistical Modeling

an image with 24 cats all purple except for one red
Uber employs statistical modeling to find anomalies in data and continually monitor data quality.

Enhanced POET: Open-Ended Reinforcement Learning through Unbounded Invention of Learning Challenges and their Solutions

Building upon our existing open-ended learning research, Uber AI released Enhanced POET, a project that incorporates an improved algorithm and allows for more diverse training environments.

Introducing Piranha: An Open Source Tool to Automatically Delete Stale Code

Uber developed Piranha to seamlessly delete code related to obsolete feature flags, leading to improved developer productivity and a cleaner codebase.

Fostering a Culture of Sponsorship: Introducing Uber’s Engineering and Sponsorship Development Program

Designed by Uber's Office of the CTO, the Engineering Sponsorship and Development Program (ESDP) pairs participants with sponsors and provides an opportunity to hone technical leadership skills.

Why We Leverage Multi-tenancy in Uber’s Microservice Architecture

Apartment building
Multi-tenancy lets Uber tag requests coming into our microservice architecture, giving us the flexibility to route requests to specific components, such as during testing scenarios.

Celebrating International Women’s Day: Meet the Women Building Uber’s Global Platform

To celebrate International Women's Day, we spoke with women from across the company whose work helps deliver impactful experiences for Uber users worldwide.

Introducing Uber’s Open Source Principles

Uber open source logo
Uber shares our principles and goals for using and contributing open source software, providing visibility into our company's approach to open source.

Under the Hood of Uber ATG’s Machine Learning Infrastructure and Versioning Control Platform for Self-Driving Vehicles

Uber ATG self-driving cars
Managing multiple machine learning models to enable self-driving vehicles is a challenge. Uber ATG developed a model life cycle for quick iterations and a tool for continuous delivery and dependency management.

Profiles in Coding: Rick Boone, Core Infrastructure, San Francisco

San Francisco skyline
Rick Boone, Strategic Advisor for Uber's Core Infrastructure group, talks about his journey from his work in site reliability to his current role in long-term planning for infrastructure health and scalability.

Tricks of the Trade: Tuning JVM Memory for Large-scale Services

latency graph
Uber engineers share their learnings on how to tune a Java Virtual Machine so as to avoid long pauses and other issues with garbage collection.

Building the Future of Mobility from the Pacific Northwest: Meet the Uber Seattle Tech Team

The Uber Seattle Tech team is responsible for building a diverse range of technologies, from developer tools to our data platform architecture.

Building a Backtesting Service to Measure Model Performance at Uber-scale

We built a backtesting service to better assess financial forecast model error rates, facilitating improved forecast performance and decision making.

Counting Calories: How We Improved the Performance and Developer Experience of UberEats.com

Screenshots from UberEats.com
Improving the performance and developer velocity for the Uber Eats web application involved a complete rewrite, developing a new architecture and using Fusion.js.

Popular Articles