Lenny's Podcast• Aug 24, 2025• 1:09:50Interview

Inside the expert network training every frontier AI model | Garrett Lord

From Lenny's Podcast

Garrett Lord•Co-founder and CEO of Handshake

Executive Summary

Handshake, a 10-year-old career marketplace for students, leveraged its existing network to launch a new AI data business.
This new venture experienced explosive growth, reaching $50M in ARR in four months and is on pace to surpass $100M in its first year, exceeding the core business's revenue.
The success is driven by a market shift in AI development, where the bottleneck has moved from pre-training on public data to post-training, which requires high-quality data from subject matter experts.
Handshake's primary competitive advantage is its proprietary network of over 18 million students and professionals, including 500,000 PhDs, which provides a zero-cost acquisition channel for expert data providers.

8 quotes

Concerns Raised

Scaling operations and hiring fast enough to meet what is described as 'unlimited demand'.
Maintaining high data quality and a positive expert experience as the platform scales rapidly.

Opportunities Identified

Capitalizing on the insatiable demand from frontier AI labs for high-quality training data.
Becoming the dominant platform for expert-driven AI data generation and post-training services.
Expanding the 'expert network' to monetize specialized skills beyond traditional employment.

Key Themes

The Shift to Post-Training in AI

Frontier AI labs have largely exhausted the gains from pre-training models on the public internet. The majority of performance improvements now come from post-training processes like RLHF and fine-tuning, which require high-quality, specialized data to enhance model capabilities in specific domains.

This creates a massive, high-value market for human-generated data, where quality and subject matter expertise are more important than sheer quantity, fundamentally changing the AI data supply chain.

Leveraging Existing Assets for New Opportunities

Handshake successfully identified that its decade-old network of students and alumni was a uniquely valuable asset for the burgeoning AI data market. They built a new, hyper-growth business on top of their existing platform, demonstrating how established companies can pivot to capitalize on technological shifts.

This serves as a powerful case study for companies to audit their existing assets—user bases, data, communities—to find an 'unfair advantage' in the AI landscape.

The Rise of the 'Expert' Economy in AI

As AI models become more sophisticated, the need for data providers has evolved from low-cost generalists to highly-skilled subject matter experts. These experts provide the nuanced data required to improve models in complex fields, creating new, high-paying gig work opportunities ($100-$200/hr for PhDs).

This trend redefines the data labeling market, making access to a diverse and scalable pool of verified experts the key competitive differentiator, rather than just process or software.

The Moat in Human-Generated Data

The speaker argues that the only sustainable moat in the human-generated AI data market is access to a large, engaged, and proprietary audience. While competitors rely on expensive acquisition channels, Handshake leverages its existing network at near-zero cost, creating a powerful and defensible business model.

This highlights that for businesses in the AI data ecosystem, community-building and network effects are more critical long-term advantages than purely technological solutions.

Incubating a Startup Within a Scale-up

To ensure focus and agility, Handshake launched its AI business as a separate, incubated unit with dedicated leadership. The CEO devoted over 80% of his time to the new venture, treating it like a new startup to navigate the zero-to-one phase effectively within a larger organization.

This provides a successful blueprint for how established companies can pursue disruptive innovation by isolating new ventures from existing bureaucracy and providing strong executive sponsorship.

Get started free

Topics

AI Data Labeling Post-Training Reinforcement Learning with Human Feedback (RLHF)Large Language Models (LLMs)Frontier AI Labs Subject Matter Experts Expert Networks Business Incubation Competitive Moats Network Effects Annual Recurring Revenue (ARR)Startup Growth Career Marketplace EdTech Future of Work

Processed Apr 3, 2026 yt-dlp + mlx-whisper + Gemini