AI Engineer

Description:

Upwind is a next-generation Cloud Security Platform that uses runtime context to identify and prioritize critical risks. Unlike traditional tools, Upwind uses runtime data to prioritize risk and inform posture, so teams can focus on what truly matters. Our eBPF-powered sensors and agentless cloud discovery deliver real-time threat protection, posture management, and API security. Together they protect cloud infrastructure end to end, from misconfigurations to malware. At Upwind, you'll have room to think creatively, explore new ideas, and make a real impact on our growth.

We're looking for a hands-on AI Engineer to join Upwind's Research team. You'll help build our core intelligence engine: LLM agents that run cloud security workflows end to end. These agents investigate runtime detections, reason over workload, identity, and cloud context, and drive remediation in production. You'll research new models and agent designs, run rigorous experiments, and ship what works as high-impact features in our platform.

What you will do

• Build LLM-based agents that investigate, triage, and remediate cloud security threats.
• Own the path from research to production: prototype new models, techniques, and agent designs, and ship the ones that work.
• Train, fine-tune, and evaluate models for security tasks using Upwind's vast data sources: runtime events, cloud logs, identity, and network activity.
• Build and extend our internal AI platform: services, pipelines, evals, and observability.
• Collaborate with Security Researchers, Product, and Engineering to turn research findings into production-grade autonomous workflows.

• 3+ years building AI/ML systems in production, including LLM-based agents built with modern AI and agent frameworks.
• Strong backend engineering skills and proficiency in Python.
• Experience training or fine-tuning ML models and LLMs.
• Strong understanding of AI system reliability: prompting and context engineering, evaluation pipelines, LLM-as-judge, regression testing, and production monitoring.
• Proven AI research ability: track the literature, frame open problems, design rigorous experiments, and benchmark new techniques against the state of the art.
• Strong ownership and the ability to work both independently and as part of a team in a fast-paced environment, bridging research and product engineering.
• Bachelor's or Master's degree in Computer Science, Data Science, or a related field.

Nice to have

• Background in cybersecurity (e.g., cloud security, threat detection, incident response).
• Hands-on experience with cloud platforms (AWS, GCP, Azure) and Kubernetes.
• 
  
• Research publications, open-source contributions, or other work that shows depth in AI.