Cluely's Roy Lee rolls up with 50 interns and 10M views in 8 days — a viral growth machine built on Instagram reels

Summary

Roy Lee's Cluely runs a growth operation built on a single mandate: get 1 billion eyeballs on the product. Everything else is secondary.

The team is 11 full-time staff with nearly 50 interns, each carrying over 100,000 followers on at least one platform. Their job is virality. Lee gives them unrestricted creative freedom to make the company go viral by any means necessary.

Cluely hit 10 million views in 8 days, driven primarily by Instagram Reels. Lee's target is 100 million views within the next month, with an eventual goal of a billion views a month. By end of summer, he plans to have 1,000 creators shipping content.

Content strategy

The winning format follows a Snapchat-style structure: a relatable hook ("I should have been a CS major"), followed by a product demo where the user reads Cluely's AI-generated interview response so stiffly that the awkwardness itself becomes the engagement driver. Lee is deliberately skeptical of AI-generated video, arguing it still falls roughly 10% short of the uncanny valley threshold needed to fool viewers. The difference between 100,000 views and 10 million is whether people believe the video is real.

Instagram audiences are nearly impossible to genuinely offend, so the play is engagement-baiting through social dynamics rather than outrage. Controversy doesn't work the same way on Instagram as it does on X.

Product direction follows content, not the other way around. Lee watches which video formats go consistently viral across different use cases, then builds the technology to match the attention. The original framing around cheating on job interviews remains the core hook, though enterprise is now described as the primary revenue driver.

Pricing and profitability

Consumer pricing is $20 a month or $100 a year. Enterprise is custom. Lee says the company is cash-flow positive and profitable without trying to be. He runs the business at high compensation deliberately, arguing that underpaying talent to protect margins is a false economy when the output is 10 million UGC views in eight days.

One large enterprise contract currently consumes most of the engineering team's bandwidth, with broader product scaling planned once that commitment is fulfilled.

Hardware

Cluely has people working on hardware in a garage, though no product details are disclosed. Lee frames hardware as a natural extension of the product's use case in real-world settings.

The core tension in Cluely's story is whether the attention machine is building toward a durable product or simply racing ahead of its own foundation. Lee's approach inverts the conventional startup sequence: go viral first, identify which use cases resonate, then build the product. That strategy is working at the metrics level. Whether a profitable, fast-growing but engineering-constrained team can convert 10 million views into something defensible before a better-resourced competitor copies the format remains unclear.