CTGT, an enterprise AI company focused on scalable model customization and deployment, has raised an oversubscribed $7.2 million seed round to accelerate its product roadmap and go-to-market strategy. Gradient, Google’s early-stage AI investment arm, led the round, with participation from General Catalyst, Y Combinator, Liquid 2, Deepwater, and an impressive group of angel investors, including François Chollet (creator of Keras), Paul Graham (Y Combinator), and Wes McKinney (creator of Pandas).
This funding round marks CTGT’s first capital raise, highlighting strong investor confidence in its ability to remove AI compute limitations and improve real-time model adaptability.
As enterprises transition from proof-of-concept AI models to full-scale production, compute and energy demands have become a major barrier. While AI applications like chatbots are relatively low-risk, high-stakes use cases in security and enterprise automation require greater reliability, adaptability, and efficiency.
"CTGT's launch is timely as the industry struggles with how to scale AI within the current confines of computing limits," said Darian Shirazi, Managing Partner at Gradient. "CTGT removes those limits, enabling companies to rapidly scale their AI deployments and run advanced AI models on devices like smartphones."
CTGT’s co-founder and CEO, Cyril Gorlla, has been studying this challenge for years. His 2023 paper presented at ICLR introduced a methodology that speeds up AI model evaluation and training by up to 500x, significantly improving accuracy. This research laid the foundation for CTGT’s platform.
The fresh capital will fuel CTGT’s product development and enterprise expansion efforts, ensuring that its AI deployment platform reaches more organizations seeking scalable, on-the-fly model training solutions.
Unlike other vendors that merely flag model issues, CTGT’s platform automatically refines and retrains AI models in real-time, eliminating the need to take them offline for updates. This capability is critical for companies that need their AI models to adapt to evolving security threats, policy changes, and shifting user behavior.
CTGT is already working with a Fortune 10 company to deploy secure, on-device AI and has multiple enterprise customers leveraging its software for AI risk management and compliance.
CTGT's platform allows enterprises to ensure AI model compliance with corporate standards in privacy, security, and operational guidelines. The system continuously refines AI performance as conditions evolve, enabling businesses to be more responsive to threats and demand fluctuations.
For example, CTGT can detect emerging security risks, such as prompt engineering attacks, and adjust AI models in real time to counteract vulnerabilities. Additionally, it can identify and correct hallucinations, factual inaccuracies, and potential data leakage, helping enterprises deploy trustworthy AI solutions.
"The lack of certainty and trust in models’ output is a significant barrier to adoption in high-stakes industries like healthcare and finance," said Gorlla. "By greatly improving accuracy, CTGT is removing that barrier."
One of CTGT’s first adopters, Ebrada Financial, used the platform to enhance the factual accuracy of customer service chatbots. "Previously, hallucinations and errors in chatbot responses led to high volumes of live support requests," said Ley Ebrada, Founder of Ebrada Financial. "CTGT has dramatically improved chatbot accuracy, reducing the need for human intervention."
