Deccan AI, a fast-growing player in the AI post-training ecosystem, has secured $25 million in a Series A round to expand its data generation, evaluation, and reinforcement learning services for advanced AI systems.
The all-equity round was led by A91 Partners, with participation from Susquehanna International Group and Prosus Ventures.
Building The Backbone Of AI Post-Training
As leading AI labs such as OpenAI and Anthropic continue developing frontier models, a growing share of the work required to refine these systems is being outsourced. This includes tasks like generating high-quality training data, evaluating outputs, and improving model performance through reinforcement learning.
Deccan AI is positioning itself at the center of this shift, offering specialized services that help make AI systems more reliable in real-world applications.
Founded in late 2024, the company supports a range of advanced use cases—from enhancing coding and agent capabilities to enabling models to interact with external systems via APIs. Its work also extends into emerging areas such as “world models,” which allow AI to better understand physical environments, including robotics and vision-based systems.
Advertisement
Serving Frontier Labs And Enterprises
Deccan AI works with major technology players, including Google DeepMind and Snowflake. The company currently serves around 10 clients while managing dozens of active projects simultaneously.
Beyond custom services, Deccan is also developing its own products, including an evaluation suite called Helix and an operations automation platform designed to streamline AI workflows.
Scaling With A Global Talent Network
Headquartered in the San Francisco Bay Area, Deccan AI operates a large delivery hub in Hyderabad and employs around 125 staff. It also leverages a vast contributor network of over one million individuals, including students, specialists, and PhD-level experts.
On a monthly basis, between 5,000 and 10,000 contributors are actively engaged in projects. A significant portion of these contributors hold advanced degrees, particularly for high-complexity assignments.
The company’s model reflects the increasing demand for domain-specific expertise in AI development, where accuracy is critical and tolerance for errors is minimal.
India As A Strategic Talent Hub
While its customer base is largely composed of U.S.-based AI companies, most of Deccan’s workforce is concentrated in India. This geographic focus allows the company to maintain tighter quality control compared to competitors that operate across dozens of countries.
The approach also highlights India’s growing role in the global AI ecosystem—not as a primary builder of frontier models, but as a key supplier of high-quality talent and training data.
At the same time, Deccan has begun expanding sourcing efforts into other regions, including the United States, to access specialized expertise in areas like semiconductor design and geospatial data.
A Rapidly Expanding Market
The market for AI training and post-training services has grown quickly alongside the rise of large language models. Companies such as Scale AI, Surge AI, Turing, and Mercor are all competing to provide data labeling, evaluation, and reinforcement learning capabilities.
However, Deccan emphasizes that quality remains one of the biggest unresolved challenges in the space. As AI systems move closer to production environments, even small errors in training data can have significant downstream consequences.
The work is also highly time-sensitive, with AI labs often requiring large volumes of high-quality data within tight deadlines—sometimes within days.
Strong Early Traction
Since launching, Deccan AI has experienced rapid growth, increasing revenue tenfold over the past year and reaching a double-digit million-dollar run rate.
The business remains highly concentrated, with around 80% of revenue coming from its top five customers—a reflection of the current structure of the frontier AI market.
With fresh capital and growing demand for high-quality AI training infrastructure, Deccan AI is positioning itself as a critical layer in the next phase of AI development.
