Inside the little-known expert network quietly training every frontier AI model | Garrett Lord (Handshake CEO)

08/24/25
Source: Lenny’s Newsletter by Lenny Rachitsky. Read the original article

TL;DR Summary of How Handshake Built a $100M AI Data-Labeling Business Leveraging Academic Experts

Handshake transformed its vast network of over 3.5 million advanced degree holders into a groundbreaking AI data-labeling powerhouse, hitting $50 million in revenue within four months. This innovative approach taps into highly specialized human expertise—like PhDs in physics and biology—to enhance AI model accuracy and performance. The company’s strategy of operating a distinct startup within a startup enabled rapid scaling and market adaptation. Rather than replacing entry-level jobs, this model creates “Iron Man suits” that amplify junior employees’ productivity tenfold.

Optimixed’s Overview: How Leveraging Academic Experts Is Revolutionizing AI Data Labeling and Startup Growth

Unlocking the Power of Expert-Led AI Training

Handshake, originally known as a career network for students and graduates, discovered a unique asset: the world’s largest network of academic experts, including 500,000 PhDs and 3 million advanced degree holders. Recognizing the critical role of expert-level data labeling in advancing AI, they launched a separate business unit focused solely on this niche.

The Importance of Expert Data Labeling

  • Specialized Knowledge: AI models require nuanced, expert-verified data to improve. For example, a biology PhD’s detailed eight-hour workday contributes significantly to training models like GPT-5.
  • Human-AI Collaboration: Experts help address AI’s limitations, ensuring continuous improvement and more reliable outputs.
  • Scalable Innovation: By creating a dedicated startup environment, Handshake rapidly scaled the business, achieving $50M revenue in four months and projecting to surpass $100M in one year.

Strategic Insights and Future Outlook

Handshake’s approach highlights a broader trend where AI doesn’t replace entry-level jobs but rather enhances them, creating “Iron Man suits” that boost junior employees’ productivity by ten times. Their model underscores the value of separating innovative projects from core operations to maintain agility and ownership. Looking ahead, the integration of expert human input with AI systems is positioned to be a key driver in the evolution of both technology and workforce dynamics.