EU remote
Applied AI Engineer - Operations
About this role
Action1 is an autonomous patch management platform trusted by many Fortune 500 companies. Cloud-native, infinitely scalable, highly secure, and configurable in 5 minutes, it just works and is always free for the first 200 endpoints, with no functional limits. By pioneering autonomous OS and third-party application patching, Action1 helps organizations eliminate routine work, reduce ransomware and security risks, and protect the digital employee experience.
It provides real-time vulnerability assessment and remediation across remote and on-site endpoints without requiring a VPN, while peer-to-peer patch distribution reduces bandwidth use on local networks. In 2026, Action1 was recognized as the fastest-growing founder-led cybersecurity company in America on the Inc. 5000. Action1 is led by industry veterans Alex Vovk and Mike Walters, American entrepreneurs who founded Netwrix, which has grown into an industry-leading, multibillion-dollar cybersecurity company.
The Role We are looking for a senior Applied AI Engineer to join Operations and improve how our business teams work. You will work directly with Support, Finance, Sales Ops, HR, IT, and Legal to turn repetitive tasks and disconnected workflows into reliable AI-powered tools and automations. We build AI to solve real problems, not to follow hype or impress investors. You will investigate how work gets done, propose improvements, and own technical decisions and delivery.
We expect you to take initiative, deliver useful results quickly, and improve solutions through short iterations and feedback. What you’ll do Work with teams to understand their workflows, find bottlenecks, and propose improvements. Agree process changes with the COO and relevant teams before implementation. Build AI tools and automations that connect business systems and handle tasks such as document processing or ticket triage.
Deliver working versions quickly and improve them through feedback. Prioritize opportunities by time saved, error reduction, and risk. Measure whether solutions deliver those benefits, reduce costs, and are actually used. Run and support what you build. Test AI outputs, keep integrations reliable as systems change, and manage failures, latency, and cost. Protect customer and employee data through limited access, minimal exposure, and redaction before model calls.