Principal AI/ML Engineer

Build the intelligence layer that transforms runtime telemetry into autonomous security. Security is becoming autonomous. RoonCyber.AI is building an AI Runtime Security platform that not only observes AI workloads but uses AI to investigate, correlate, explain, and respond to security events. We're looking for an exceptional AI/ML engineer to architect the intelligence powering next-generation security operations. What You'll Own Security AI Copilot Detection Tuning and Recommendation Engine Autonomous Investigation Agents Evidence Correlation Engine Natural Language Security Interface MCP and AI Tool Integrations Security Knowledge Graph Enrichment Response Recommendation Engine AI Compliance Automation Model Evaluation and Guardrails Problems You'll Solve Turn millions of runtime events into meaningful investigations. Use AI to correlate runtime telemetry, identities, vulnerabilities, and attack paths. Develop autonomous agents that investigate incidents and collect evidence. Enable CISOs to ask natural language questions about AI workloads. Technologies Python LLM APIs (OpenAI, Anthropic, Bedrock, Vertex AI) LangGraph, CrewAI, AutoGen MCP Vector databases Knowledge graphs PyTorch, Hugging Face, LangChain Kubernetes and cloud-native AI infrastructure What Great Looks Like Build production AI investigation agents. Deliver AI-assisted detection engineering. Reduce SOC investigation time through autonomous workflows. Provide explainable AI recommendations for security analysts. We're Probably Looking For You If You've built production AI systems. You enjoy reasoning over complex security data. You care about explainability and trust. You like inventing new AI workflows. Ideal Background Anthropic, OpenAI, Databricks, NVIDIA, Microsoft AI, Google DeepMind, AWS Bedrock, Protect AI, Lakera, HiddenLayer. Experience building enterprise AI applications rather than research prototypes. What We Don't Care About Publishing research papers. Winning Kaggle competitions. Training foundation models from scratch. Academic credentials over practical engineering. The Interview Design an AI-powered investigation agent. Explain how you would evaluate hallucinations, accuracy, and trust. Demonstrate how AI should assist—not replace—security analysts. Why Join Define the future of AI-native security operations. Build technology that will fundamentally change how SOC teams investigate attacks. Work directly with founders on category-defining technology. Meaningful ownership and equity. Engineers Who Will Love This Job You love applying AI to difficult engineering problems. You enjoy experimenting with emerging agent frameworks. You think AI should augment expert analysts. You want to build products, not demos. This Probably Isn't For You If You only want to build chatbots. You prefer academic research over production engineering. You dislike rapidly evolving AI technologies. You prefer maintaining existing ML pipelines. As a fully remote company, we’re looking for candidates who take initiative, communicate proactively, and are comfortable working independently while maintaining close overlap with US Eastern Time hours. Show more Show less