Datacurve Raises $15 Million to Take on Scale AI

As the AI industry matures, the competition for high-quality data has intensified, paving the way for companies like Mercor, Surge, and Scale AI, led by Alexandr Wang. However, with Wang transitioning to run AI at Meta, many investors see an opportunity and are now funding new companies with innovative strategies for collecting training data.

One of these emerging companies is Datacurve, a Y Combinator graduate focused on providing high-quality data specifically for software development. On Thursday, Datacurve announced a $15 million Series A round, led by Mark Goldberg at Chemistry, with participation from employees at DeepMind, Vercel, Anthropic, and OpenAI. The Series A follows a $2.7 million seed round backed by former Coinbase CTO Balaji Srinivasan.

Datacurve's Unique Datacurve's tacurve uses an innovative "bounty hunter" system to engage killed software engineers to contribute to difficult-to-source datasets. The company compensates contributors for their efforts, having distributed over $1 million in bounties so far.

However, co-founder Serena Ge emphasises that the company's core advantage is financial. According to Ge, for high-value services like software development, compensation for data work is typically far lower than in conventional employment. Instead, the company's most crucial goal is to provide a positive user experience.

"We treat this as a 'consumer product, not a data labelling operation," Ge explained. "We "spend a lot of "time thinking about: How can we optimise it so that the people we want are interested and get onto our platform?"

The Growing Need for Sophisticated Data Collection

The demand for post-training data is increasing as AI models become more complex. Early models were trained using simpler datasets, but today's AI products operate in complex reinforcement learning (RL) environments. These environments must be strategically built through specific and targeted data collection.

As the need for both data quantity and quality grows, companies like Datacurve, which specialise in high-quality data collection, have a distinct advantage. Ge believes that Datacurve's data, currently focused on software engineering, can be applied to other industries such as finance, marketing, and even medicine.

"What we're doing right now're we're creating an infrastructure for post-training data collection that attracts and retains highly competent people in their own domains," Ge said.

Looking "head

Datacurve's $15 million round sets the company on a path to expand its operations and refine its data-collection model. With the evolving needs of AI products, Datacurve's specialised data from talented individuals positions it as a promising player in the highly competitive AI space.