Key facts
- Uber's CTO embedded top AI engineers with finance, legal, and HR teams.
- The initiative, called "agentic pods," involved 30 AI-proficient engineers.
- AI agents were developed to automate tasks requiring access to multiple systems.
- Financial pacing reports now take 10 minutes, down from two days.
- Capital allocation across 150 cities now takes 30 minutes, down from 15 hours.
- Uber plans to form a dedicated team to scale this AI integration.
Uber is implementing a new strategy for AI integration by embedding its top AI engineers within various business departments, including finance, legal, and human resources. This approach, termed "agentic pods," involves engineers working directly with employees to understand and automate complex tasks.
According to Uber's CTO Praveen Neppalli Naga, this method has led to significant efficiency gains. For instance, financial pacing reports that previously took two days can now be completed in just 10 minutes. Similarly, the task of allocating capital across Uber's 150 operating cities has been reduced from 15 hours to 30 minutes.
Naga emphasized that effective automation requires understanding how work is actually performed, rather than relying solely on process diagrams. The company has already run 16 such pods over the past two months.
Despite these efficiencies, Uber, like many tech companies, has been increasing its AI spending. However, Chief Operating Officer Andrew Macdonald noted in May that justifying the expenditure on AI has become more challenging, as the spending has not yet translated into a proportional increase in "useful" consumer features.
Uber intends to continue and expand its use of the agentic pod model, with plans to form a dedicated team to further scale the initiative and fundamentally redesign business operations using AI.