WorkThe story, in brief

‘Tokenmaxxing’ Starts to Fade as Companies Eye Agentic Coding Costs

Nobody is talking about the shift away from token optimization. Salesforce and Uber are already betting on agentic coding—and the ROI math is completely different.

Illustration of independent geometric mechanisms passing paper tasks along branching amber tracks.
AI agents and the coordination of work.AI illustration by KeyNews
The KeyNews take

Why it matters

The industry is moving past 'tokenmaxxing' (optimizing for token efficiency) toward agentic AI workflows, forcing enterprises to rethink how they measure AI ROI. This signals a strategic pivot in how Fortune 500 companies are allocating AI spend and evaluating success.

The key facts

4 to know
  1. Salesforce and Uber investing heavily in agentic coding approaches

  2. Industry debate emerging on ROI measurement for agentic vs. token-optimized models

  3. Shift from token efficiency optimization to agentic workflow capabilities

  4. Enterprise decision-making moving toward agent-based AI deployment metrics

Go to the source

Newcomer (VC)newcomer.co

Publisher excerpt: Salesforce & Uber are among those investing heavily. A fierce debate is roiling the industry on when & how to measure ROI.
Read original report
Back to today's editionMore work news

Keep reading

Related stories

More from Work