![Uber logo](https://blog.uber-cdn.com/cdn-cgi/image/width=2160,quality=80,onerror=redirect,format=auto/wp-content/uploads/2023/11/Rider_App_Icon_iOS-1.png)
![Featured image for Presenting the Engineering Behind Uber at Our Technology Day](https://blog.uber-cdn.com/cdn-cgi/image/width=2160,quality=80,onerror=redirect,format=auto/wp-content/uploads/2017/04/UberTechDayHeader_-1024x439.jpg)
On March 10, 2017 in Palo Alto, Uber hosted the inaugural Uber Technology Day, sponsored by LadyEng, Uber’s community of female engineers, data scientists, and allies founded in 2014. The event brought together people from across Uber’s engineering organizations—as well as dozens of external engineers from the Bay Area—for a day of in-depth presentations, lightning tech talks, and conversations around Uber’s engineering challenges and culture, as well as those of the general tech industry.
“Through LadyEng and other initiatives, Uber strives to increase the number of women in its workforce as well as place them in some of the most critical roles, giving them key responsibilities and thereby creating equal opportunity in the truest sense,” said Komal Mangtani, a senior engineering director and co-lead of the event.
![Image](https://blog.uber-cdn.com/cdn-cgi/image/width=1024,quality=80,onerror=redirect,format=auto/blog/wp-content/uploads/2017/04/image15.jpg)
Presentations highlighted technology from across the organization and technology stack, including self-driving vehicle development, mapping technology, and the engineering involved throughout the life of an Uber trip.
![Image](https://blog.uber-cdn.com/cdn-cgi/image/width=1024,quality=80,onerror=redirect,format=auto/blog/wp-content/uploads/2017/04/image06.jpg)
“When LadyEng and Uber began planning the conference over a year ago in early 2016, we set out with a big goal: to shine a spotlight on Uber and the engineering challenges we tackle daily,” said Denali Lumma, the engineering manager for our People Engineering team. “While Uber’s founding mission is to make transportation as reliable as running water, the work that we do has many more industry-defining applications. Uber Technology Day presented an opportunity to distill our work and highlight the amazing people who do it.”
Uber Technology Day Presentations
Here are several of the day’s presentations, also available on the Uber Engineering YouTube Channel:
The Life of an Uber Trip
Fran Bell, a data science manager with teams developing Intelligent Decision Systems and our Forecasting Platform (including anomaly detection), described what goes on behind the scenes from ride request through end of trip, describing the technical aspects that enable each stage of a ride to take place seamlessly.
Monorepo to Multirepo and Back Again
Aimee Lucido, a mobile engineer on Uber’s Driver Signups team, spoke about Uber’s transition from a monorepo to multirepo and back again, discussing her team’s experience growing the structure of an Android app at scale. “When your app is small, it makes sense to put everything in one place,” said Lucido. “But as your team grows and your app follows suit, engineers must make investments for long-term success.”
Automatic Algorithm Selection for Anomaly Detection: From Prototype to Production
Yiren Lu, a New York City-based software engineer on Uber’s Observability Operations team (who started off as a data science intern back in Summer 2014), presented on how Uber is automating anomaly detection through a recent project called Metric Reliability Layer (MeRL) to better determine whether an algorithm will track bugs, outages, and other high tech hiccups with a high probability of accuracy.
Mobile Architecture for Scale
Jennifer Chen, a mobile engineer on Uber’s Intelligent Dispatch team, spoke to the audience about how we overhauled the Uber rider app last year to accommodate additional features, including scheduled rides and accelerators for one tap requests, to make the experience smoother for users. Our mobile teams rewrote the the app in six months, making the software more reliable and scalable than ever before.
Building an Experimentation Platform at Uber
Eva Feng, a data scientist on Uber’s Data Science Platform team, shared her thoughts on what makes Uber’s Experimentation Platform unique to the industry. While Uber launches new features to users worldwide, it also maintains real-time experiment health monitoring for stability and security. In her presentation, Eva outlined some of the biggest accomplishments of her team: “In the past year, we…helped the Customer Relationship Management (CRM) team to identify which process is the best for onboarding drivers and referring more riders on our app,” said Feng. “Secondly, we improved the reliability and safety of the app through a staged rollout project. And thirdly, we discovered a lot of impactful mobile experiments to identify pickup hotspots and other key information.”
How Does Uber Handle New Year’s Eve?
Janani Narayanan, a software engineer on the Marketplace Fares team, gave the audience a bird’s eye view into how Uber manages New Year’s Eve and other wildly popular dates for riders and drivers by preventing outages before they have a chance to occur. With zero outages experience worldwide on New Year’s Eve 2016, Janani and her team worked on Uber’s traditionally busiest night of the year to identify and fix performance issues in real-time.
How the Uber Developer Platform Makes the Future Possible
Charlyn Gonda, a developer advocate with the API Partnerships team, discussed how the Uber Developer Platform enables Uber engineers and the wider developer community to make the future possible. During her talk, Gonda gave attendees a look inside one particular API, Trip Experiences, to learn more about some of the futuristic experiences developers can build with Uber today.
Thinking Like a Human: Toward Flexible Artificial Intelligence
Noah Goodman, a fellow at Uber AI Labs and an associate professor of psychology, computer science, and linguistics at Stanford University, discussed some of the often overlooked but very important “ingredients” in modern machine learning: deep learning classification systems. “These deep learning classification systems that are trying to do things like look at a picture and decide whether it’s a cat or a boat have improved radically in the last six years,” said Goodman. “Back in 2010, it was nothing to write home about, especially compared to humans. As of last year, however, these systems are at or beating human-level performance.”
Building a Scalable, Reliable Data Platform
Engineering manager Deepti Chheda and engineer Ayesha Yasmeen, both with the Data Platform team, discussed their experience building a data platform capable of handling Uber’s massive scale of data while still delivering high quality, reliable results using Streamific and other tools. “Using this system, we were able to ingest from thousands of streams across multiple data centers in near real time,” said Chheda. “This meant that regardless of if trips were happening in Europe or in Southeast Asia, we were able to get the raw data of all those trips into our data warehouse almost instantaneously.”
Explore IPv6 Deployment at Uber
Jean He, a network engineer with Uber’s Tech Ops team, spoke about about our recent IPv6 adoption and Uber’s new physical infrastructure. Her presentation showcased how her team was rolling out new features in the IPv6 industry production architecture, while raising awareness of the value of IPv6 deployment among the Internet community. “The IPv6 deployment is never a zero or all implementation. It requires testing incrementally and iteratively,” said He. “We also want to advocate engineers’ awareness when they are writing code to make sure that they are able to support IPv6.”
Driving into the Future
Attendees also heard technical talks from members of the Uber Advanced Technology Group (ATG), including Claire Delaunay, an engineering director and Otto co-founder, and Joan Wang and Poornima Kaniarasu, software engineers. During their presentations, the engineers discussed topics related to the state of self-driving technologies, including object-oriented detection, the opportunities and challenges facing the development of “robot cars”, and the future of machine learning in automated driving.
![Image](https://blog.uber-cdn.com/cdn-cgi/image/width=1024,quality=80,onerror=redirect,format=auto/blog/wp-content/uploads/2017/04/image05-1.jpg)
During her presentation on developing “robot cars” at Uber, Claire discussed ATG’s first commercial delivery in October 2016 during which their prototype vehicle drove 120 miles in Colorado fully autonomously. “The day you actually see the truck on the road after you’ve done all of these experiments, it’s crazy honestly,” said Delaunay. “When you see a robot driving like that, it’s just so steady.”
Sharing a Leadership Perspective
After the presentations, moderator and mobile engineer Paulina Ramos hosted a panel on the intersection of diversity, culture, and management in Uber Engineering among members of Uber’s leadership team, including Chief Technology Officer Thuan Pham, Director of Engineering for Technology Services Shobhana Ahluwalia, engineering manager Denali Lumma, and Chief Security Officer Joe Sullivan.
![Image](https://blog.uber-cdn.com/cdn-cgi/image/width=1024,quality=80,onerror=redirect,format=auto/blog/wp-content/uploads/2017/04/image07.jpg)
The panel discussed how their diverse backgrounds in technology and other industries have informed their approach to innovative problem-solving at all levels of engineering. According to Ahluwalia, a huge component of fostering the ability to solve these problems at a company like Uber is mentorship. “Sometimes you’re not in a position to define a mentor. At the end of the day, a mentor is anybody who is willing to give you time, who is invested in making you a better you and is ready to hold you accountable,” she said. “Mentoring is a process of always learning and always self-evaluating, and every interaction is something you give to and something you take away from.”
The conference closed with a Q&A session between CEO Travis Kalanick and the audience, moderated by Mangtani.
Kalanick began his discussion by referencing a viral image of a bronze statue of a young girl defiantly facing the famous “Charging Bull” statue on New York City’s Wall Street. Opening with this picture facilitated a transparent conversation about Uber’s culture, and Kalanick stressed the need for Uber and other technology companies to remove “the bulls” that prevent women engineers from excelling and growing in the workplace.
![Image](https://blog.uber-cdn.com/cdn-cgi/image/width=1024,quality=80,onerror=redirect,format=auto/blog/wp-content/uploads/2017/04/image11.jpg)
Looking Back, Moving Forward
Uber Technology Day was the first official technology symposium held by Uber Engineering, as well as the first of many opportunities to present our projects at all stages of development.
“Over the past several years, LadyEng has really come into its own as a defining element of the Uber community at all levels and across teams,” said Tasneem Minadakis, a senior engineering manager on Uber’s Driver Onboarding team. “While LadyEng has always worked behind the scenes to help grow our organization, the event gave our members a chance to showcase their work not only for the broader engineering team at Uber, but also members of the wider Silicon Valley engineering community with a stake in these technologies.”
The Uber Engineering Blog will feature many of these presentations over the course of the next several months through individual blog posts, starting with a look at technology behind the Uber Development Platform next week.
Want to get involved? If the work outlined above interests you, consider applying for a role on our engineering team, and if you would like to attend future Uber Engineering events, join our Meetup group.
![Stephanie Romo](https://blog.uber-cdn.com/cdn-cgi/image/width=2160,quality=80,onerror=redirect,format=auto/wp-content/uploads/2022/08/sromo.jpeg)
Stephanie Romo
Stephanie Romo is the program manager for Uber's LadyEng group.
Posted by Stephanie Romo
Related articles
Most popular
![Post thumbnail](https://blog.uber-cdn.com/cdn-cgi/image/width=2160,quality=80,onerror=redirect,format=auto/wp-content/uploads/2024/04/background_top-1024x585.png)
DragonCrawl: Generative AI for High-Quality Mobile Testing
![Post thumbnail](https://blog.uber-cdn.com/cdn-cgi/image/width=2160,quality=80,onerror=redirect,format=auto/wp-content/uploads/2024/05/MA-logo.png)
From Predictive to Generative – How Michelangelo Accelerates Uber’s AI Journey
![Post thumbnail](https://blog.uber-cdn.com/cdn-cgi/image/width=2160,quality=80,onerror=redirect,format=auto/wp-content/uploads/2024/05/2048px-Categorization_system_2-1024x683.jpeg)
DataK9: Auto-categorizing an exabyte of data at field level through AI/ML
![Post thumbnail](https://blog.uber-cdn.com/cdn-cgi/image/width=2160,quality=80,onerror=redirect,format=auto/wp-content/uploads/2023/02/Uber_Ring_HCV_Large_1-1024x576.png)
Public transport agencies trial mixed fleets to implement local on-demand transport
Products
Company