
Ex-RapidAPI prodigy Iddo Gino launches AI startup Datawizz with $12.5M Seed round
The 26-year-old entrepreneur, ousted from his first unicorn in 2023, returns with a startup aiming to slash the costs of running AI models.
Iddo Gino, once hailed as a “wonder child” for founding RapidAPI at the age of 17 and later turning it into a unicorn before being ousted in 2023, is back with a new venture. His latest company, Datawizz, launched earlier this year with just four employees, has already raised $12.5 million, a relatively large sum for a Seed-stage round.
Iddo Gino: "AI is currently a pyramid scheme. When everything is revealed, we will be ready"
Full list of Israeli high-tech funding rounds in 2025
The round was led by the U.S. venture capital firm Human Capital. Funds 91VC and BGV, both of which invested in RapidAPI, also participated. Gino, who has been living in the United States for the past few years, founded the startup there but plans to build R&D operations in Israel following the raise and begin recruiting locally.
Datawizz is developing technology aimed at reducing the costs of using artificial intelligence. “Companies spent $8.4B on API calls to LLMs in just the first half of 2025—more than double the figure for all of 2024,” said Gino, Founder and CEO of Datawizz. “The pace of adoption is incredible. The challenge is that many applications have upside down unit economics and deliver inconsistent results. Datawizz fixes this by replacing the massive generic models with smaller, Specialized Language Models that deliver faster, cheaper, and more accurate results.”
Related articles:
According to the company, organizations can deploy dozens of specialized models, each significantly smaller than a standard large language model, and train them for highly specific tasks. Datawizz’s system routes each query to the most relevant specialized model, turning to a general model only when no specialization is available. The company predicts that by 2030, around 70% of AI traffic will be handled by specialized models with fewer than 5 billion parameters.