4 ms·

> software powered by machine learning models leverages data to predict rider supply and driver demand; and data scientists use data to improve machine learning

by mbesto 8y ago

> software powered by machine learning models leverages data to predict rider supply and driver demand; and data scientists use data to improve machine learning models for better forecasting. Why does any of that require real-time analysis? What's the business need for this being instantaneous? Feels very NIH...

treypitt 8y ago
national institute of health?
- IanCal 8y ago
  Not Invented Here - refers to people avoiding bringing in something already built from outside and building in-house instead. https://en.wikipedia.org/wiki/Not_invented_here https://en.wikipedia.org/wiki/Not_invented_here
jon37 8y ago
Most businesses don't require real time forecasting, but I would speculate that Uber likely uses it to decide when and where to activate surge pricing.
nemothekid 8y ago
A friend, who works for Uber, was once explaining the difference between AirBnb’s tech stack and Uber’s even though they are both “marketplaces” In short, Uber has a ton more infrastructure because they need to do more in real time. Calculating the price of a trip from point A to point B needs to be done in real time. Then there are other factors like Pool, surge, and various rider preferences that all need to be accounted for. Uber is sort of unique in how much of their stack nessicates real time analytics
- mbesto 8y ago
  I don't disagree but the use cases Uber is talking about is not app functionality but for their analytics team (from the article): - Build dashboards to monitor our business metrics - Make automated decisions (such as trip pricing and fraud detection) based on aggregated metrics that we collect - Make ad hoc queries to diagnose and troubleshoot business operations issues None of those require real-time information, because there isn't real-time decision-making required from the information presented.* With the exception of the "automated decisions", then I don't see the need.

by mbesto 8y ago

treypitt 8y ago

national institute of health?

IanCal 8y ago
Not Invented Here - refers to people avoiding bringing in something already built from outside and building in-house instead. https://en.wikipedia.org/wiki/Not_invented_here https://en.wikipedia.org/wiki/Not_invented_here

jon37 8y ago

Most businesses don't require real time forecasting, but I would speculate that Uber likely uses it to decide when and where to activate surge pricing.

nemothekid 8y ago

A friend, who works for Uber, was once explaining the difference between AirBnb’s tech stack and Uber’s even though they are both “marketplaces” In short, Uber has a ton more infrastructure because they need to do more in real time. Calculating the price of a trip from point A to point B needs to be done in real time. Then there are other factors like Pool, surge, and various rider preferences that all need to be accounted for. Uber is sort of unique in how much of their stack nessicates real time analytics

mbesto 8y ago
I don't disagree but the use cases Uber is talking about is not app functionality but for their analytics team (from the article): - Build dashboards to monitor our business metrics - Make automated decisions (such as trip pricing and fraud detection) based on aggregated metrics that we collect - Make ad hoc queries to diagnose and troubleshoot business operations issues None of those require real-time information, because there isn't real-time decision-making required from the information presented.* With the exception of the "automated decisions", then I don't see the need.