
"To win in security, companies will need to deploy security-specific models in products optimized for real security workflows. To build these models, you need specialized data, domain-specific evaluation, and deep expertise in post-training. Our team is one of the few in security able to do that."
"Dfs-mini1 was trained in security-specific environments and can generalize beyond smart contracts, meaning that the training approach transfers across security domains."
Depthfirst, a software security startup, has secured $80 million in Series B funding, increasing its total funding to $120 million within three months. The investment was led by Meritech Capital, with support from various venture firms. Founded in 2024 by leaders from DeepMind and Databricks, Depthfirst focuses on software and infrastructure security. The company launched Dfs-mini1, an in-house security model aimed at securing cryptocurrency smart contracts. The funds will be used to expand the AI research team and develop additional security models.
Read at SecurityWeek
Unable to calculate read time
Collection
[
|
...
]