How Axur scales its proprietary AI models without worrying about infrastructure
|
“With Azion, we’ve been able to scale our proprietary AI models without worrying about infrastructure. These solutions inspect millions of websites daily, detect and neutralize threats with speed and precision, and execute the fastest automatic takedown in the market.” Fabio Ramos, CEO at Axur |
|
Impact at Scale
- 5 minutes from detection to takedown request.
- 30k+ automated takedowns monthly.
- 450 million new websites added to verification monthly.
Challenges
Axur faced several challenges before adopting Azion’s AI Inference:
- Managing and scaling infrastructure: High costs and considerable engineering effort required with managed-infrastructure.
- Cost and speed: Faster response times while saving on operational costs.
- Maintaining precision: Ensuring the accuracy of security analyses was critical, yet challenging, given the increasing sophistication of cyber threats.
Solution
Axur’s innovation on AI enables them to reduce takedown times by enhancing decision-making processes for detecting and responding to online threats. This enhancement also improved service delivery, allowing Axur to overcome the limitations of traditional managed-infrastructure and become more competitive with smarter, more efficient AI solutions.
By adopting Azion’s AI Inference, Axur had several significant benefits:
- Efficiency: Replaced managed-infrastructure with Azion AI Inference’s serverless architecture, where you only pay for what you really use.
- Dynamic scaling: Automated end-to-end deployment and scaling of advanced AI models on Azion distributed infrastructure, reducing manual intervention.
- Custom models: With LoRA, Axur can adapt the models available on Azion’s serverless platform.
- Security as a priority: Freed up teams to focus on improving threat detection and response, not on infrastructure maintenance.
Results
- Smooth system integration: Azion’s compatibility with existing AI frameworks ensured a fast transition, minimizing disruptions during the deployment of AI Inference.
- Performance and Scale: The scale and performance of the infrastructure provided by AI Inference allowed Axur to quickly scale while delivering the same levels of performance and service availability.
- Cost efficiencies: By leveraging Azion’s fully-managed infrastructure, Axur reduced its reliance on traditional SaaS and cloud computing, leading to savings and faster response times.
About Axur
Axur is a cybersecurity company specializing in digital risk protection and cyber threat intelligence. Founded in 1999, Axur is a leading SaaS platform for monitoring, detecting, and eliminating online threats that pose risks to organizations’ security and reputation. Their solutions are recognized for their advanced automation, leveraging AI and large language models (LLMs) to prioritize risks, automate phishing and fraud takedowns, and provide real-time threat intelligence. As of 2024, Axur employs over 200 people and reported annual revenues of US$24.9 million.
Axur’s platform is designed to empower security teams to manage threats beyond the traditional perimeter, offering tools for external attack surface management, credential exposure detection, and deep & dark web monitoring. Their technology aggregates and analyzes massive volumes of data—adding 15 million new websites to their data lake daily—and uses GenAI to detect phishing, counterfeit products, and brand impersonation across global marketplaces and digital channels.