AI Trainer Job Board Hits $1,360 in June Revenue After Google De-Index

The founder launched aitrainer.work in January 2024 as a simple aggregator pulling AI training, RLHF, and data annotation jobs from various platforms into one searchable database. The site also includes platform intelligence and user reviews, a free AI training academy, and a growing talent pool. Under the hood, about a dozen scrapers run regularly against major job platforms, consolidating fresh opportunities into a single dataset.

The business model relies on referral programs embedded in the platforms the site covers. When a candidate applies through aitrainer.work's link and gets onboarded or hired, the site earns a commission or revenue share on their earnings. However, the mechanics revealed a critical delay: platforms onboard hundreds of people but often lack enough projects immediately available, leaving candidates waiting in the pipeline—sometimes for months, sometimes indefinitely. This meant early revenue came primarily from onboarding bonuses rather than from actual revenue share once people started working. The founder noted this was supposed to flip once candidates were assigned projects, but the lag persisted through May.

In the first six months, revenue and traffic told a lumpy story. January launched at 9.5k sessions and $709 revenue. February dropped to 7.2k sessions and $305 after a major Google update de-indexed several pages. March plummeted to $24 as the pipeline gap hit hard. April recovered slightly to 9.6k sessions and $270 as the founder launched the talent pool and free academy. May saw 12k sessions but only $150 revenue, still trapped in the pipeline bottleneck. By June, sessions climbed to 15k and revenue jumped to $1,360—driven by the newly-launched paid certification program.

The Google penalty proved less damaging than it could have been. Immediately after shipping in early January, Google ranked the site well. The February update de-indexed content, but instead of collapsing, traffic remained steady because Bing and ChatGPT began sending users—creating an ironic situation where an AI training job board drew much of its organic traffic from AI chatbots. The founder was not deeply concerned, viewing it as a natural lifecycle for a brand-new site, and focused instead on long-form content and different approaches rather than emergency Google recovery.

A major inflection arrived when one of the platform partners reached out directly. They needed high-quality candidates for specific RLHF roles. Using the talent pool of 500+ registered candidates, the founder manually sent targeted invites to relevant prospects. Based on the data, this achieved an 8 percent placement rate—described as huge for the industry. This moment revealed the true value proposition: the aggregator was not just a traffic arbitrage play, but a sourcing tool that employers would pay for if it could deliver quality talent consistently.

The technical stack evolved quickly. The founder initially used OpenGraph's Ampcode free tier, then switched to Claude Pro in March, treating it as a co-development partner. The founder handled product, design, and strategy while Claude questioned ideas, co-designed plans, and wrote the bulk of the code. This partnership worked smoothly, though API costs spiked unexpectedly when the founder hit a weekly limit and blew twenty dollars in half an hour mid-sprint.

A UX-first design philosophy became the site's competitive moat. While most sites delegate design to AI, the founder conducted research and iterated based on user feedback. This defensibility mattered when competitors copied the entire site—including the academy modules and paid certification—without changing even the names. But the copies built "AI sloppy" and lacked engaged users, while the original continued to iterate around real feedback and pull ahead.

Looking forward, the founder's roadmap reflects lessons learned. The obvious near-term play is fixing Google SEO by embedding proprietary data onto each job board page, ensuring they offer value beyond just surfacing jobs found elsewhere. Longer-term, the vision includes proprietary tools like a CV builder (instead of relying on affiliate ads), a B2B hiring portal where companies pay for direct access to the talent pool (described as "real MRR"), and proprietary market research based on aggregated data to generate authority and backlinks. For now it remains a one-person operation running on passion and a solid hourly rate equivalent, but the unit economics are finally pointing in the right direction.

The site's core insight—that AI training job seekers need a unified aggregator because contract work is inherently multi-platform—addresses a real friction point. Unlike traditional employment job boards, these opportunities are temporary projects that demand continuous hunting; a candidate cannot rely on a single platform's pipeline. By consolidating opportunities from a dozen scrapers running against major platforms, aitrainer.work removes the need to tab-switch between sites.

The revenue model's early bottleneck reveals a structural challenge in the AI training gig economy: platforms onboard hundreds but lack immediate work, leaving candidates in a months-long waiting state. This created a lag between onboarding bonuses (the founder's main early revenue) and revenue share from actual work. The shift in June toward paid certification suggests the founder is diversifying away from this dependency—a strategic pivot grounded in observing that the referral-based model alone cannot sustain rapid growth.

The 8 percent placement rate from the B2B sourcing request is particularly significant: it demonstrates that the founder's manual curation and UX-first design create a quality signal that employers value. This opens a path to recurring revenue (the "real MRR" mentioned) by inverting the business model—charging companies for direct access to vetted talent rather than relying solely on affiliate commissions. The founder's willingness to compete on design and user research rather than AI-generated sites has created defensibility: competitors who copied the site's structure still lack an engaged user base.