Upwind Security develops cloud security software for organizations running multi-cloud infrastructures. Its platform brings together cloud security posture management, runtime vulnerability management, API security, and related capabilities to help DevOps and Security teams detect threats in real time, prioritize vulnerabilities, and manage infrastructure and application risk more efficiently.
AI Engineering Lead for LLM Agents – Hybrid in Tel Aviv
Lead the development of production-grade LLM agent workflows and backend infrastructure for cloud security operations. Build reliable AI systems that investigate, reason about, and respond to complex security events.
Description
- Create and maintain the orchestration layer for autonomous agents handling complex security operations
- Help develop shared infrastructure for models, data pipelines, training, evaluation, and observability
- Develop the agentic workflow engine, including tool use, state management, retries, and evaluation
- Build backend services and APIs that deliver AI capabilities securely and at scale
- Partner with AI researchers to turn research results into production-ready autonomous workflows
- Take ownership of the reliability, observability, and performance of production agentic systems
Requirements
- Mid- to senior-level engineering experience or relevant technical military experience
- Strong backend engineering expertise focused on machine learning and AI systems
- Practical experience building and deploying LLM applications and agentic frameworks
- Ability to work effectively in a fast-paced setting that connects research with product engineering
- Experience with LangChain, LangGraph, or a comparable agent-orchestration framework is a significant advantage
- Experience using evaluation frameworks and observability tools for non-deterministic AI systems