@UberEng profile picture
Uber Engineering @UberEng
Joined June 2014
  • Tweets 11,540
  • Following 5,394
  • Followers 58,871
1,536 Photos and videos
Accessing data doesn’t have to be so hard!  Discover how Uber supports simple yet powerful reporting experiences on top of complex data sets in this blog about building a data abstraction layer. 🔗 uber.com/us/en/blog/data-abs… #UberEngineering #UberEngandSci #DataEngineering

Simplifying Data and Product Integrations with a Data Abstraction Layer

Accessing data doesn’t have to be so hard! Discover how Uber supports simple yet powerful reporting experiences on top of complex data sets in this blog about building a data abstraction layer.

uber.com
1
9
2,173
Are you at #CVPR2026? 🗻Drop by the @Uber AV Labs booth (#647) to learn more about our work accelerating the AV ecosystem. Learn more about our AV Labs team here: uber.com/global/en/av-labs/
2
1
15
6,175
How do you protect data at scale? Uber’s File Semantic Analyzer uses GenAI to semantically classify and summarize file content, boosting security and cutting false positives. Check it out in our latest blog: uber.com/us/en/blog/ubers-fi… #UberEngineering #AI #SecurityEng

Building a File Semantic Analyzer: Guarding Outbound Data at Scale with AI

How do you protect data at scale? Uber’s File Semantic Analyzer uses GenAI to semantically classify and summarize file content, boosting security and cutting false positives.

uber.com
2
11
2,271
Learn how Uber modernized artifact storage using a managed cloud platform and a proxy to reduce operational risk, control egress costs, and scale build infrastructure. 🔗 uber.com/us/en/blog/moderniz… #UberEngineering #DistributedSystems #Caching #Modernization

Modernizing Artifact Storage at Uber

Learn how Uber modernized artifact storage using a managed cloud platform and a proxy to reduce operational risk, control egress costs, and scale build infrastructure.

uber.com
3
13
2,208
AI agents can take powerful actions, but can you prove who acted and why? Here’s how Uber built agent identity and end-to-end provenance for secure, auditable tool calls. 🔗 uber.com/us/en/blog/solving-… #UberEngineering #AgenticAI

Solving the Identity Crisis for AI Agents

AI agents can take powerful actions, but can you prove who acted and why? Here’s how Uber built agent identity and end-to-end provenance for secure, auditable tool calls.

uber.com
1
9
44
13,566
Learn how Uber deployed a deep transformer model with graph data pipelines to solve real-time traffic forecasting, improving route quality and arrival times for millions of customers worldwide. Check out our latest blog: uber.com/us/en/blog/scaling-… #UberEngineering #DeepLearning

Scaling Real-Time Traffic Forecasting with a Graph-Aware Transformer

Learn how Uber deployed a deep transformer model with graph data pipelines to solve real-time traffic forecasting, improving route quality and arrival times for millions of customers worldwide and...

uber.com
1
4
1,669
What do scheduling TV ads, shipping crates, and Uber incentives have in common? They’re all multiple knapsack problems. See how we engineered a scalable solution with our new system, Tarot, in our latest blog. 🔗 uber.com/blog/solving-multip… #UberEngineering #MachineLearning

Beyond Prediction: Solving the Multiple Knapsack Problem at Scale: How Uber Optimizes Incentives

What do scheduling TV ads, shipping crates, and Uber incentives have in common? They’re all multiple knapsack problems. See how we engineered a scalable solution with our new system, Tarot.

uber.com
6
1,384
Moving millions of compute cores to Kubernetes without dropping a single request was no small feat. Go behind the scenes on the @KubernetesPod with Uber Staff Engineer Lucy Sweet (@int_underflow). 👇 Link below. #UberEngineering #Kubernetes
Kubernetes Podcast from Google@KubernetesPod
May 13
This episode we speak to Lucy Sweet. Staff Engineer @Uber and lead for the @kubernetesio Node Lifecycle WG. Lucy shares Uber 5 Millions cores on #k8s story at scale and why it's always DNS 😃 📃 kubernetespodcast.com/episod… 🔊 kubernetespodcast.com/subscr…
2
10
1,937
What if every PM had a fast first-pass reviewer before a PRD reached the review room? We built an AI PRD Evaluator at Uber to surface blind spots early, uncover prior artifacts, and give product reviews higher signal. Read more: uber.com/us/en/blog/first-pa… #Uber #AIForPMs

Lessons from Building a First-Pass AI PRD Reviewer at Uber

What if every PM had a fast, context-rich first-pass reviewer before a PRD reached the review room? We built an AI PRD Evaluator at Uber to surface blind spots early, uncover prior artifacts, and...

uber.com
1
1
3
1,810
With over 2 million cores, 1% efficiency is worth millions. See how Uber tuned Go's runtime stack to boost CPU utilization for a service by 10%. Read More: uber.com/blog/zero-growth-st… #Uber #UberEngineering #StackAllocation

Zero-Growth Stack, Real Gains: How Stack Allocation Can Save 10% CPU in Go

With over 2 million cores, 1% efficiency is worth millions. See how Uber tuned Go's runtime stack to boost CPU utilization for a service by 10%.

uber.com
1
11
2,250
Discover how Uber Engineering uses Ansible® to automate configuration management for its global enterprise network. Read More: uber.com/us/en/blog/ansible-… #Uber #UberEngineering

How Ansible® Automation Powers the Uber Corporate Network at a Global Scale

Discover how Uber Engineering uses Ansible® to automate configuration management for its global enterprise network.

uber.com
8
1,790
Last week, Uber Engineering hosted a Seattle Systems meetup. We explored topics like analyzing code merge safety at scale and discussed the physical limits of the internet.🌟 It was great to see Seattle's vibrant systems engineering community in action! 🙌 #UberEngineering
3
1,263
Uber Engineering retweeted
Praveen Neppalli
@praveenTweets
Apr 23
Excited to bring Uber & Uber Eats into Claude! Browse restaurants, check fare ranges, see ETAs, and complete trips or orders seamlessly. A strong example of AI simplifying everyday use. Great team effort to make it fast, reliable, and intuitive. 🚀 @UberEng
44
31
467
61,520
Discover how Uber AI built an ML validation system to automatically catch critical ID swaps and position jumps in video bounding box annotations, streamlining data quality for object tracking. Read More: uber.com/us/en/blog/validati… #Uber #UberEngineering #MachineLearning

Validating Bounding Box Annotations

Discover how Uber AI built an ML validation system to automatically catch critical ID swaps and position jumps in video bounding box annotations, streamlining data quality for object tracking.

uber.com
3
1,516
It was a big week for Uber Engineering at hashtag #ICSE2026. Read the award-winning research and learn how we’re scaling our testing infrastructure below. AutoCover: Scaling Test Coverage: conf.researchr.org/details/i… AI-Driven Mobile Chaos Engineering: arxiv.org/pdf/2602.06223
1
3
10
1,713
Meet Odin’s hybrid core allocation: CPU when you need it, elastic sharing when you don’t—built for efficient scaling. Read More: uber.com/blog/hybrid-core-al… #Uber #UberEngineering #Odin

Hybrid Core Allocation: From Overallocation to Reliable Sharing

Meet Odin’s hybrid core allocation: CPU when you need it, elastic sharing when you don’t—built for efficient scaling.

uber.com
1
1
12
2,537
Uber Eats scaled its homefeed discovery engine by transitioning to transformer-based generative recommenders and a modernized ML infrastructure to deliver real-time, personalized recommendations to millions. Read More: uber.com/us/en/blog/next-gen… #UberEngineering #MachineLearning

Next-Gen Restaurant Recommendation with Generative Modeling and Real-Time Features

Learn how Uber Eats scaled its homefeed discovery engine by transitioning to transformer-based generative recommenders and a modernized ML infrastructure to deliver real-time, personalized recommen...

uber.com
1
14
2,104
Discover how Uber uses OpenSearch™’ gRPC™/Protobuf APIs for modern search and ingestion, and learn about our contributions to the gRPC/Protobuf transport in the OpenSearch project. Read More: uber.com/us/en/blog/high-per… #Uber #UberEngineering #OpenSearch

Accelerating Search and Ingestion with High-Performance gRPC™ in OpenSearch™

Discover how Uber uses OpenSearch™’ gRPC™/Protobuf APIs for modern search and ingestion, and learn about our contributions to the gRPC/Protobuf transport in the OpenSearch project.

uber.com
1
14
3,541
Explore how we optimized Petastorm to cut deep learning training time by 6× and eliminated hidden randomness at Uber. Read More: uber.com/us/en/blog/accelera… #Uber #UberEngineering #MachineLearning

Accelerating Deep Learning: How Uber Optimized Petastorm for High-Throughput and Reproducible GPU...

Explore how we optimized Petastorm to cut deep learning training time by 6× and eliminated hidden randomness at Uber.

uber.com
3
1,774
Load more