Crescendo Bought a 3,000-Person BPO
Crescendo charges per resolved case, not per hour
AI Natives
Companies built on AI agents from day one, and how hard they can hit existing industries.
Periodic Labs: Automating the Reading of Results First
How lab results become training data for the company's own model
Jooba: One Dedicated AI Agent for Every Role
How per-role AI agents and people split the work of recruiting
A CSO That Never Touches Data: Stanford's Experiment
A study that models a drug R&D organization on 11 AI agents
Prism: Listing Each Candidate's Weaknesses Too
How Prism finds hiring criteria in past hires and candidate feedback
Lila: Scientist Review Shrinks to a Safety Check
How Lila divides experiments among models, instruments, and people
Norm Law: The Client Layer That Stays When Lawyers Leave
How Norm Ai builds the agents and Norm Law supervises them
Knowlee: Agents Make the Cards, Only People Move Them
How Knowlee works with agents and a review queue
Edison's Kosmos: 57.9% of Interpretations Held Up
How Kosmos divides research work, and why people focus on verification
Harper: Agents Read the Paperwork, People Close the Deal
How Harper divides the work of insurance brokerage between agents and people
CREAO: A Company of About 25, Coordinated by Agents
How agents coordinate and people decide, with no middle managers
Sakana's Manual Audit Caught What the Evaluator Missed
How Sakana redesigned the division of labor between scientists and agents
Crosby: Training Agents on Lawyers' Edits
How Crosby splits contract review with agents and charges per contract
Cognition Starts Work at the Pull Request
How a team builds its own software with Devin, its coding agent
Recursive Built a Model in Under Four Months
How an AI research loop built a model, and lessons for R&D teams
